i

UNIVERSIDADE FEDERAL DE MINAS GERAIS

Programa de Pós-Graduação em Engenharia Metalúrgica, Materiais e de Minas

Dissertação de Mestrado/Master of Science Research Report

Autor/Author: Dalila Chaves Sicupira

Orientador/Supervisor: Prof. Marcelo Borges Mansur

“Remoção de manganês de drenagem ácida de mina

utilizando carvão de osso.”

“Removal of manganese from acid mine drainage

using bone char.”

February/2012.

i

UNIVERSIDADE FEDERAL DE MINAS GERAIS

Programa de Pós-Graduação em Engenharia Metalúrgica, Materiais e de Minas

Dalila Chaves Sicupira

“Remoção de manganês de drenagem ácida de mina

utilizando carvão de osso.”

“Removal of manganese from acid mine drainage

using bone char.”

Dissertação de Mestrado apresentada ao Programa de Pós-Graduação em Engenharia

Metalúrgica, Materiais e de Minas da Universidade Federal de Minas Gerais

Área de concentração: Tecnologia Mineral

Orientador: Prof. Marcelo Borges Mansur

Belo Horizonte

Escola de Engenharia da UFMG

February, 2012.

ii

ACKNOWLEDGEMENTS

Initially I would like to thank God for the grace to get here.

My mother, for love, force and unconditional support. To my brother, my boyfriend and

all my family, those have always believed in me.

Prof. Marcelo Borges Mansur the opportunity, friendship and learning.

My colleagues Ana Cláudia, Ana Paula, Carolina, Douglas, Eduardo, Emily, José

Alberto, Joyce, Leandro, Marcela, Marcos, Michele, Ronie, Sara and Thiago, for

companionship and friendship.

People from Laboratory of corrosion, chemical analysis and Metallography.

Financial support from FAPEMIG, CNPq, CAPES and INCT-Acqua: National Institute

of Science and Technology of Mineral Resources, Water and Biodiversity.

Bone Char do Brasil Ltda as well as Prof. Sônia Denise Ferreira Rocha

(DEMIN/UFMG) and Prof. Ana Cláudia Ladeira (CDTN) for the fruitful discussion and

contribution.

Techs. Ilda, Isabel, Andréia, Cláudia and Patrícia.

Profs. Dagoberto Brandão, Vicente Buono, Edwin Auza, Virgínia Ciminelli, Wander

Luiz Vasconcelos, Paulo Roberto Brandão, Adriana França, Leandro Oliveira and Maria

Silvia for the analysis.

PPGEM: Cida and Nelson.

Ultimately, I thank all those who directly or indirectly contributed to the realization of

this work.

iii

SUMMARY

LIST OF FIGURES .............................................................................................................. v

LIST OF TABLES .............................................................................................................. vii

RESUMO ............................................................................................................................ viii

ABSTRACT ......................................................................................................................... ix

1. INTRODUCTION ......................................................................................................... 1

1.1. References .............................................................................................................. 3

2. OBJECTIVES ................................................................................................................ 5

2.1. Main objective ............................................................................................................ 5

2.2. Specific objectives ..................................................................................................... 5

3. BACKGROUND ........................................................................................................... 6

3.1. Acid mine drainage .................................................................................................... 6

3.2. Manganese .................................................................................................................. 7

3.3. Technologies for AMD treatment ............................................................................. 8

3.3.1. Manganese removal by adsorption/ion exchange ........................................... 11

3.4. Bone char .................................................................................................................. 14

3.5. Adsorption ................................................................................................................ 16

3.5.1. Adsorption isotherms ........................................................................................ 17

3.5.2. Adsorption kinetics ........................................................................................... 18

3.5.3. Fixed bed adsorption ......................................................................................... 20

3.6. Desorption ................................................................................................................ 23

3.7. References ................................................................................................................ 24

4. BATCH REMOVAL OF MANGANESE FROM ACID MINE DRAINAGE

USING BONE CHAR ........................................................................................................ 32

Abstract ................................................................................................................................ 32

4.1. Introduction .............................................................................................................. 33

4.2. Experimental ............................................................................................................ 34

4.2.1. Reagents and instrumentation........................................................................... 34

4.2.2. Manganese removal studies .............................................................................. 35

4.3. Results and discussion ............................................................................................. 36

4.3.1. Characterization analysis .................................................................................. 36

4.3.2. Batch sorption kinetics...................................................................................... 39

iv

4.3.3. Batch sorption isotherms .................................................................................. 44

4.3.4. Mechanism for the removal of manganese with bone char ............................ 47

4.4. Conclusions .............................................................................................................. 52

4.5. Acknowledgements .................................................................................................. 53

4.6. References ................................................................................................................ 53

5. ADSORPTION OF MANGANESE FROM ACID MINE DRAINAGE EFFLUENTS

USING BONE CHAR: CONTINUOUS FIXED BED COLUMN AND BATCH

DESORPTION STUDIES. ................................................................................................. 56

Abstract ................................................................................................................................ 56

5.1. Introduction .............................................................................................................. 57

5.2. Experimental ............................................................................................................ 58

5.2.1. Reagents ............................................................................................................. 59

5.2.2. Instrumentation .................................................................................................. 60

5.2.3. Continuous fixed bed column runs .................................................................. 60

5.3. Results and Discussion ............................................................................................ 62

5.3.1 Fixed bed sorption runs...................................................................................... 62

5.3.2 Desorption studies .............................................................................................. 69

5.4. Conclusions .............................................................................................................. 70

5.5. Acknowledgments .................................................................................................... 72

5.6. References ................................................................................................................ 72

6. FINAL CONSIDERATIONS AND SUGGESTIONS FOR FUTURE WORKS ...... 75

v

LIST OF FIGURES

Figure 3.1: Active and passive treatment technologies for AMD remediation (Adapted

from Johnson and Hallberg, 2005). ...................................................................................... 9

Figure 3.2: Breakthrough curve for adsorption in a fixed bed (Adapted from Reynolds

and Richards, 1995). ........................................................................................................... 22

Figure 4.1: X-ray pattern of bone char............................................................................... 37

Figure 4.2: FTIR spectra of bone char using KBr discs ................................................... 37

Figure 4.3: Morphology and chemical composition of bone char by SEM-EDS: (a, b)

bone char as received, and (c, d) bone char after contact with AMD effluent. ............... 38

Figure 4.4: Effect of solid/liquid ratio on the kinetics of manganese removal with bone

char from (a) laboratory solution (C0 = 100 mg L-1

), and (b) AMD effluent (25°C, 150

rpm, 417-833 µm, continuous curves are pseudo-second order model). ......................... 41

Figure 4.5: Effect of particle size on the kinetics of manganese removal with bone char

from (a) laboratory solution (C0 = 100 mg L-1

), and (b) AMD effluent (25°C, 150 rpm,

2/400 g mL-1

, 417-833 µm, continuous curves are pseudo-second order model). .......... 43

Figure 4.6: Effect of initial pH on the isotherm of manganese removal from AMD

effluent with bone char (25°C, 150 rpm, 53 µm, 0.25/100 g mL-1

, 48 hours; continuous

lines are Langmuir model). ................................................................................................. 45

Figure 4.7: Effect of temperature on the isotherm of manganese removal with bone char

from (a) laboratory solution (C0 = 100 mg L-1

), and (b) AMD effluent (150 rpm, < 53

µm, 1/400 g mL-1

, 48 hours; continuous lines are Langmuir model). ............................. 46

Figure 4.8: Relationship between release of calcium into solution and manganese

adsorbed to bone char (25°C, 150 rpm, 417–833 µm, 0.25/100 g mL-1

, pHi = 5.76, C0 =

100 mg L-1

, 24 hours)......................................................................................................... 49

Figure 4.9: Intraparticle mass transfer for manganese removal with bone char from (a)

laboratory solution (C0 = 100 mg L-1

), and (b) AMD effluent (25°C, 150 rpm, 2/400 g

mL-1

)..................................................................................................................................... 51

Figure 5.1: Effect of bed height on the breakthrough of manganese removal with bone

char in a continuous fixed bed column (T = 25±1ºC, 417-833 µm): (a) Laboratory

solution (flow rate = 7.5 mL min-1

, C0 = 100 mg L-1

, pHi = 5.76), (b) AMD effluent

(flow rate = 3.0 mL min-1

, pHi = 2.96, continuous curves are Thomas model). ............. 63

vi

Figure 5.2: Morphology of bone char by SEM (scanning electron microscopy): (a) bone

char as received, and (b) bone char after contact with AMD effluent in the column. .... 66

Figure 5.3: Effect of flow rate on the breakthrough of manganese removal with bone

char in a continuous fixed bed column using the laboratory solution (C0 = 100 mg L-1

,

pHi = 5.76, bed height = 8 cm, T = 25±1ºC, 417-833 µm)............................................... 67

Figure 5.4: Effect of initial pH on the breakthrough of manganese removal with bone

char in a continuous fixed bed column using the AMD effluent (flow rate = 3.0 mL min-

1, bed height = 8 cm, T = 25±1ºC, 417-833 µm, continuous curves are Thomas model).

.............................................................................................................................................. 69

vii

LIST OF TABLES

Table IV.1: Pore distribution for the particle size range studied……………………….36

Table IV.2: Chemical characterization of the AMD effluent. .......................................... 39

Table IV.3: Parameters of pseudo-second order model for the effect of solid/liquid ratio

(25°C, 150 rpm, 417-833 µm, pHi = 5.5-5.7, C0 = 100 mg L-1

in the laboratory solution).

.............................................................................................................................................. 42

Table IV.4: Parameters of pseudo-second order model for the effect of particle size

(25°C, 150 rpm, 2/400 g mL-1

, pHi = 5.5-5.7, C0 = 100 mg L-1

in the laboratory

solution). .............................................................................................................................. 44

Table IV.5: Parameters of Langmuir equation for the pH effect for AMD effluent (25°C,

150 rpm, <53 µm, 0.25/100 g mL-1

, 48 hours). ................................................................. 45

Table IV.6: Parameters of Langmuir equation for the effect of temperature (150 rpm,

pHi = 5.5-5.7, < 53 µm, 1/400 g mL-1

, 48 hours). ............................................................. 47

Table IV.7: Parameters of intraparticle diffusion model for manganese adsorption on

bone char (25°C, 150 rpm, 2/400 g mL-1

, pHi = 5.5-5.7, C0 = 100 mg L-1

in the

laboratory solution). ............................................................................................................ 50

Table V.1: Chemical characterization of the AMD effluent (Sicupira et al., 2012). ...... 60

Table V.2: Parameters of Thomas model for the adsorption of manganese onto bone

char (T = 25ºC, 417-833 µm). ............................................................................................ 65

Table V.3: Desorption of manganese from loaded bone char using various elution

solutions (500 mg, 100 mL, 25°C, 150 rpm, 417-833 µm)............................................... 70

viii

RESUMO

Um estudo em batelada e em leito fixo foi realizado com o objetivo de avaliar a

viabilidade do uso de carvão de osso na remoção de manganês presente em efluentes de

drenagem ácida de mina (DAM). Ensaios com solução sintética contendo apenas

manganês também foram realizados para fins de comparação. Testes de equilíbrio

revelaram que a capacidade máxima de adsorção de manganês baseada na equação de

Langmuir foi de 22 mg g-1

para o efluente e 20 mg g-1

para a solução sintética. A

cinética de adsorção de manganês em carvão de osso foi descrita pelo modelo de

pseudo-segunda ordem. A remoção de manganês foi influenciada pelas variáveis

operacionais razão sólido/líquido e pH da fase aquosa, sendo favorecida em valores de

pH próximos da neutralidade. O efeito do tamanho das partículas e da temperatura foi

não significativo para a faixa investigada. A difusão intrapartícula revelou-se a principal

etapa limitante para a adsorção de manganês, mas a difusão da camada limite também

pode afetar sua remoção em partículas menores. As seguintes variáveis operacionais

foram avaliadas nos testes contínuos em leito fixo: massa de carvão de osso, vazão e pH

inicial. Variações significativas na resistência à transferência de massa de manganês no

carvão de osso foram identificadas utilizando o modelo de Thomas. Efeito significativo

da massa de adsorvente foi observado apenas em testes com a solução

sintética. Nenhuma mudança significativa no volume de ruptura foi observada para

diferentes vazões. Aumentando-se o pH inicial de 2,96 para 5,50, o volume de trespasse

foi aumentado. A capacidade máxima de adsorção de manganês calculado em testes

contínuos foi 6,03 mg g-1

para o efluente e 26,74 mg g-1

para solução sintética. O estudo

incluiu também testes de dessorção utilizando soluções de HCl, H2SO4 e água, mas

resultados não promissores foram obtidos devido aos baixos níveis de dessorção

associados a elevada perda de massa. Apesar disso, a remoção de manganês de efluentes

de DAM usando carvão de osso como adsorvente é tecnicamente viável, atendendo a

legislação ambiental. É interessante notar que utilizando carvão de osso para remover o

manganês pode-se evitar a necessidade de correção do pH do efluente pós-tratamento

com cal. Além disso, ele pode remover íons flúor e outros metais, sendo um material

alternativo bastante interessante para tal aplicação.

Palavras-chave: manganês; carvão de osso; drenagem ácida de mina; coluna de leito

fixo; adsorção/dessorção.

ix

ABSTRACT

Batch and continuous fixed bed tests were carried out aiming to evaluate the feasibility

of using bone char for the removal of manganese from acid mine drainage

(AMD). Tests with laboratory solution containing solely manganese at typical

concentration levels were also carried out for comparison purposes. Equilibrium tests

revealed that Langmuir based maximum manganese uptake capacity was 22 mg g-1

for

AMD effluent and 20 mg g-1

for laboratory solution. Manganese kinetics onto bone char

was best described by the pseudo-second order model. Manganese removal was

influenced by operating variables solid/liquid ratio and pH of the aqueous phase, and it

was favored at nearly neutral pH values. The effect of particle size and temperature was

not significant for the operating range investigated. It has been found that intraparticle

diffusion is the main rate-limiting step for manganese sorption systems but additional

contribution from boundary layer diffusion might also affect removal when bone char

particles of smaller sizes are used. The following operating variables were evaluated for

continuous fixed bed tests: mass of bone char, flow rate and initial pH. Significant

variations in resistance to the mass transfer of manganese into the bone char were

identified using the Thomas model. A significant effect of bed height was observed only

in tests with laboratory solution. No significant change on the breakthrough volume was

observed with different flow rate. Increasing the initial pH from 2.96 to 5.50, the

breakthrough volume was increased. The maximum manganese loading capacity

calculated in continuous tests using bone char for AMD effluent was 6.03 mg g-1

and

for laboratory solution was 26.74 mg g-1

. The study included also desorption tests using

solutions of HCl, H2SO4 and water but no promising results were obtained due to low

desorption levels along with relatively high mass loss. Despite of this, the removal of

manganese from AMD effluents using bone char as adsorbent is technically feasible

thus attending environmental legislation. It is interesting to note that the use of bone

char for manganese removal may avoid the need for pH correction of the effluent after

treatment with lime. Also, it can remove fluoride ions and other metals, so bone char is

a quite interesting alternative material for the treatment of AMD effluents.

Key-words: manganese; bone char; acid mine drainage; continuous fixed bed column;

adsorption/desorption.

1

1. INTRODUCTION

Given its dynamics and persistence, acid mine drainage (AMD) is one of the most

serious environmental problems of mining industry. Such phenomenon occurs when

pyrite and other sulfide minerals are oxidized due to its exposure to oxygen and water,

producing sulfuric acid and dissolved metals (Nascimento, 1998). The main sources of

AMD are open pit or underground mines, waste rock piles, tailings storage and ore

stockpiles. The raise in the concentration of metal ions in the water due to AMD is an

important source of contamination of watercourses, especially when ions can be spread

to the food chain. Aiming to minimize the severity of such impacts, environmental

protection laws require appropriate plans for mine closure as an attempt to increase the

cycle life of closed mines including steps of decommissioning and recovery of degraded

areas (Gonçalves, 2006).

A significant increase in the concentration of metal ions in water due to AMD

represents an important source of contamination of watercourses, especially when ions

can be spread to the food chain. In addition, many wastes containing heavy metals have

high contamination power, which can easily reach reservoirs and rivers that are sources

of water supply for the cities (Gonçalves, 2006).

In Brazil, the problem of acid mine drainage has been highlighted in coalfields in the

south of the country and in the Industrial Mining Complex of Poços de Caldas (CIPC),

Nuclear Industries of Brazil, INB. In this latter area there are two watersheds, Ribeirão

das Antas and Rio Verde. The waters of this region are mainly used for agricultural

irrigation and watering cattle, and also the practice of fishing. Within a radius of 20 km,

is not observed any domestic use of these waters (Gonçalves, 2006). Currently, the

industrial complex remains disabled and in process of decommissioning.

In CIPC, a large quantity of waste consisted mainly of uranium was produced and it was

disposed in large areas surrounding the mined area, called “bota-fora”. Until July 1996,

CIPC had processed 2.111.920 tons of ore (dry basis) which resulted in the generation

of 44.56 million m3 of wastes from mining (Nascimento, 1998).

2

These areas becames a source of acid mine drainage containing radionuclides (uranium,

thorium and radium) and stable elements (manganese, zinc, fluoride, iron, etc.) at

concentration levels above those allowed by Brazilian legislation for direct discharge

into the environment (Gonçalves, 2006; CONAMA, 2005). The current treatment of

acid waters consists of metals precipitation with lime, but manganese ions are

notoriously difficult to remove from AMD due to their complex chemistry (Bamforth et

al., 2006; Robinson-Lora and Brennan, 2010a). For complete precipitation, pH around

11 is required, which makes the consumption of lime to be around 350 tons per month

(Gonçalves, 2006).

Therefore, new technologies to minimize the environmental impacts of mineral

processing operation are urged. Among them, metals adsorption process has been

investigated (Korn et al., 2004; Soto et al., 2005; Bosco et al., 2005), however relatively

little attention has been directed to manganese removal. The use of adsorption in

hydrometallurgical processes has many advantages, since it allows the recovery of

metal ions from very dilute solutions and it has also the ability to process large volumes

of solutions where others operations would be unfavorable. In this context, the study of

manganese removal by adsorption process is justified, because the environmental

impact of the CIPC can be reduced with the advantage to drastically reduce the lime

consumption as well.

Bone char can be successfully applied for metals removal (Pan et al., 2009; Choy et al.,

2004). This material has been used as an alternative material to remove several metals,

such as copper, zinc, cadmium, arsenic, mercury, etc (Choy et al., 2004; Chen et al.,

2008; Cheung et al., 2000; Hassan et al., 2008), however the use of bone char in the

manganese removal has little information until date.

In the view of the aforementioned, the aim of this work is to evaluate the feasibility of

using bone char in the treatment of AMD containing manganese in order to reach the

levels required by the Brazilian legislation for direct discharge into the environment

(CONAMA, 2005) which limits the total dissolved Mn concentration in the effluent to 1

mg L-1.

3

1.1. References

ANNADURAI, G., LING, L.Y., LEE, J.F., 2008. Adsorption of reactive dye from

solution by chitosan: isotherm, kinetic and thermodynamic analysis. Journal of

Hazardous Materials, 152, 337-346.

BAMFORTH, S.M., MANNING, D.A.C., SINGLETON, I., YOUNGER, P.L.,

JOHNSON, K.L., 2006. Manganese removal from mine waters - investigating the

occurrence and importance of manganese carbonates. Applied Geochemistry, 21, 1274-

1287.

BOSCO, S.M.D., JIMENEZ, R.S., CARVALHO, W.A., 2005. Removal of toxic metals

from wastewater by Brazilian natural scolecite. Journal of Colloid and Interface

Science, 281, 424-431.

CHEN, Y.-N., CHAI, L.-Y., SHU, Y.-D., 2008. Study of arsenic(V) adsorption on bone

char from aqueous solution. Journal of Hazardous Materials, 160, 168-172.

CHEUNG, C.W., PORTER, J.F., MCKAY, G., 2000. Sorption kinetics for the removal

of copper and zinc from effluents using bone char. Separation and Purification

Technology, 19, 55-64.

CHOY, K.K.H., KO, D.C.K., CHEUNG, C.W., PORTER, J.F., MCKAY, G., 2004.

Film and intraparticle mass transfer during the adsorption of metal ions onto bone char.

Journal of Colloid and Interface Science, 271, 284-295.

CONSELHO NACIONAL DO MEIO AMBIENTE. Resolução CONAMA n.º

357/2005: Classificação dos corpos de água e diretrizes ambientais para o seu

enquadramento, bem como estabelece as condições e padrões de lançamento de

efluentes. Brasília, 2005. Disponível na Web em:

http://www.mma.gov.br/port/conama/index.cfm.

GONÇALVES, C.R., 2006. Remoção de manganês e recuperação de urânio presentes

em águas ácidas de mina. M.Sci. Thesis, CDTN, Belo Horizonte, Brazil.

4

KORN, M.G.A., SANTOS JR., A.F., JAEGER, H.V., SILVA, N.M.S., COSTA, A.C.S.,

2004. Cooper, zinc and manganese determination in saline samples employing FAAS

after separation and preconcentration on Amberlite XAD-7 and Dowex 1X-8 loaded

with alizarin Red. Journal of Brazilian Chemical Society, 15(2), 212-218.

NASCIMENTO, M.R.L., 1998. Remoção e recuperação de urânio de águas ácidas de

mina por resina de troca iônica. M.Sci. Thesis, UFSCar, São Carlos, Brazil.

PAN, X., WANG, J., ZHANG, D., 2009. Sorption of cobalt to bone char: kinetics,

competitive sorption and mechanism. Desalination, 249, 609-614.

ROBINSON-LORA, M.A., BRENNAN, R.A., 2010a. Biosorption of manganese onto

chitin and associated proteins during the treatment of mine impacted water. Chemical

Engineering Journal, 162, 565-572.

SOTO, O.A.J.; TOREM, M.L.; TRINDADE, R.B.E. Palygorskite as a sorbent in the

removal of manganese (II) from water mine effluents. In: XIII INTERNATIONAL

CONFERENCE ON HEAVY METALS IN THE ENVIRONMENT, Rio de Janeiro -

RJ, 2005.

5

2. OBJECTIVES

2.1. Main objective

Evaluation of the performance of bone char to treat AMD effluents containing

manganese in order to reach the level required by the environmental legislation (i.e., 1

mg L-1

according to CONAMA, 2005) for direct discharge into the environment.

2.2. Specific objectives

Physical and chemical characterizations of the bone char before and after

treatment of AMD using analytical techniques such as X-ray diffraction, Fourier

transform infrared spectroscopy, Scanning electron microscopy and Braunauer,

Emmet e Teller method.

Study of kinetics of manganese removal and evaluation of the effects of

operating parameters solid/liquid ratio and size of particles.

Evaluation of the isotherms of manganese removal in batch tests and evaluation

of the effect of operating parameters pH and temperature.

Investigation of metal desorption in order to evaluate the ability of sorbent to be

recycled and/or the metal to be recovered for reuse.

Execution of continuous experiments in laboratory columns in order to evaluate

the effect of flow rate, mass of bone char and initial pH.

6

3. BACKGROUND

3.1. Acid mine drainage

Among the main environmental aspects and impacts of mining activities, the one

associated with the contamination of surface and ground waters by acid mine drainage

(AMD) is probably one of the most significant (Nascimento, 1998). AMD may occur

when the metal or mineral of interest is associated with sulfides like pyrite (FeS2), for

example, in the extraction of gold, coal, copper, zinc and uranium. Mining wastes when

exposed to oxidizing conditions in the presence of water may result in AMD, therefore

the proper disposal of mining wastes from such operations is fundamental to prevent

and minimize the generation of AMD (Gonçalves, 2006). Besides waste rock piles, acid

mine drainage may also occur in open or underground pit mines, tailings storage and ore

stockpiles.

The basic reactions that describe the phenomenon of pyrite oxidation are:

2Fe 2 s 2 2 2 4 (3.1)

2 2 2 (3.2)

Fe 2 s 2 2

(3.3)

6 2 (3.4)

According to reaction (3.1), pyrite reacts with oxygen in the presence of water. The

ferrous ion then slowly oxidized (reaction 3.2) resulting in ferric iron. The ferric iron

formed accelerates the pyrite oxidation according to reaction (3.3), thus increasing the

acidity and the concentration of Fe2+

that enter in the cycle by reaction (3.2). The

precipitated iron hydroxide formed by reaction (3.4) is a reservoir of soluble Fe3+

(Nascimento, 1998).

7

AMD often contains high concentrations of SO42-

ions and the high acidity may

promote the dissolution of metals such as Zn, Cu, Cd, As, Fe, U, Al and Mn from the

minerals (Bamforth et al., 2006; Robinson-Lora and Brennan, 2009).

Three different processes are required for the treatment of AMD (Robinson-Lora and

Brennan, 2009):

The addition of a neutralizing agent;

The reduction of sulfate concentrations;

The removal of dissolved metals.

3.2. Manganese

The industrial applications of manganese included steel alloy, dry cell battery, glass and

ceramic, paint, ink, dye and fertilizers (Mohan and Chander, 2006a). Manganese is

found in acid mine drainage (AMD) derived from coal and metal mining, and their

concentration can vary considerably, from below 1 mg L-1

to hundreds of mg L-1

in

untreated AMD (Robinson-Lora and Brennan, 2010a).

Although it is typically found at lower concentrations in AMD and it has a lower

toxicity than most of other metal contaminants, Mn still affects the color, taste, and odor

of water (Robinson-Lora and Brennan, 2010a).

Manganese is an important trace element for the functioning and activation of many

enzymes (manganese superoxide dismutase, kinases, decarboxylases, among others) in

the human body, but at high levels, it causes damage to the brain, liver, kidneys and

nervous system (Silva et al., 2010). Because of this, discharges from mining activities to

surface water in Brazil must comply with the Brazilian environmental legislation

(CONAMA, 2005), which limits the total dissolved Mn concentration in the effluent to

1 mg L-1

.

In natural systems, manganese can form a number of different solid phases. The main

factors that affect the composition of minerals containing Mn are the pH and the Eh of

the aqueous system. In an oxidizing environment, Mn oxides are commonly formed,

8

including Mn oxyhydroxides and phyllomanganates (Lind and Hem, 1993). Mn sulfides

such as alabandite are expected to form in reducing environments, at high pH (Bamforth

et al., 2006). Manganese carbonate minerals such as rhodochrosite (MnCO3) and

kutnahorite (CaMn(CO3)2) have been identified in mine water-impacted environments

(Lind and Hem, 1993). Lind and Hem (1993) also suggest that the formation of Mn

carbonates in contaminated mine waters may be an essential mechanism for the long-

term immobilization of Mn.

The solubility of manganese in water is favored by reducing conditions (Soto et al.,

2005). In the reduced Mn2+

state, Mn is relatively soluble as MnSO40

(aq) complex at

least up to pH 8. At higher pH, it can precipitate as MnCO3 (rhodochrosite) and

Mn(OH)2 (at low CO2 pressure). In contrast, under oxidizing conditions, Mn3+

and Mn4+

species are relatively insoluble in the form of MnO2, Mn2O3, Mn3O4 and related

compounds. The removal of Mn from AMD requires either high pH or strong oxidizing

conditions (Rose et al., 2003a). The neutralized solution, however, requires pH

adjustment to the value below 9.0 (CONAMA, 2005) to be discharged.

Rose et al. (2003a) showed that if Fe2+

is present in solution, Mn cannot precipitate at

near-neutral pH. For Mn to precipitate, the water must be well oxygenated to the extent

that essentially all Fe is insoluble.

3.3. Technologies for AMD treatment

A number of active and passive treatment technologies (Figure 3.1) have been

suggested to remove contaminants from AMD. Although Mn removal is difficult,

relatively little attention has been directed to Mn removal systems (Rose et al., 2003b).

Active treatment systems are those that use mechanical energy to promote mixing of

neutralizing agents with AMD, for example, water treatment stations that use stirred

tanks (Johnson and Hallberg, 2005). Typically, active treatment of AMD involves

chemical processes by the addition of an alkaline agent and/or sulfide to promote the

precipitation of the metals as hydroxides, carbonates, or sulfides (Robinson-Lora and

Brennan, 2009). The alkaline reagents, normally used in the conventional neutralization

9

of pH of AMD (CaO, CaCO3, NaOH, Na2CO3) produce a considerable volume of solid

waste rich in iron and other metals, whose disposal represents another environmental

problem and additional costs. Other limitations of this process can be pointed out, as

lack of selectivity of the precipitation process, their constant demand for supplies,

energy and maintenance and low efficiency in the remediation of dilute solutions (Laus

et al., 2005; Robinson-Lora and Brennan, 2010b).

Aziz and Smith (1992) have studied the removal of manganese from water by

precipitation. At a final pH of 8.5 (initial pH 5.5-9.0), limestone removed 95% of

manganese, crushed brick 82%, gravel about 60% and the removal with no solid media

was less than 15%. Their results indicated that rough solid media and the presence of

carbonate are beneficial to the precipitation of manganese in water.

However, due to the limitations of active treatment, alternative passive treatments have

been investigated. In passive systems the treatment is promoted by passing the effluent

through stationary devices (wetlands, channels and drains) where the neutralizing agents

are placed (abiotic) or where the biochemical treatment is performed (biotic). Passive

biological systems can be performed in wetlands classified as aerobic or anaerobic

(Johnson and Hallberg, 2005).

Figure 3.1: Active and passive treatment technologies for AMD remediation (Adapted

from Johnson and Hallberg, 2005).

Treatment

technologies

Active

Passive

Aeration and lime addition

Limestone channels and drains

Aerobic wetlands

Anaerobic wetlands

Permeable reactive barries

10

A number of researchers have reported experiments and treatment systems in which Mn

bearing solutions flow through limestone beds. Experiments comparing a limestone bed

with a gravel bed (aluminosilicate rock) showed that the limestone bed was much more

effective in removing manganese (Aziz and Smith, 1996).

The most common method using a limestone bed is the patented Vail and Riley system,

also named the “Pyrolusite” ystem, in which a bed of limestone is inoculated with Mn-

oxidizing bacteria. Effective Mn removal requires oxidizing well-aerated water, as well

as prior removal of essentially all dissolved Fe and Al, and pH above about 6.5. Most of

the Pyrolusite systems removed manganese from influent values of 6 to 30 mg L-1

to

effluent levels between 0.5 to 1.5 mg L-1

for a period of 2 years or more, but several of

the systems have failed because of plugging of the inlet area with silt, leaves, Fe and/or

Al precipitates, grass and other materials (Rose et al., 2003b).

Wetlands consist of a complex ecosystem represented by the interaction between

terrestrial and aquatic systems that can be used to remediate AMD. Commonly known

as biological filters, wetlands have emerged as a viable option for helping to solve a

wide range of environmental and water quality problems. Research on utilization of

wetlands for the treatment of AMD to remove contaminants has been carried out in the

past decade. Heavy metals removal from AMD in wetlands is done by physical,

chemical and various biological processes which involve sedimentation, settling,

filtration, adsorption, precipitation, co-precipitation into insoluble compounds (Sheoran

and Sheoran, 2006). Even though manganese readily precipitates, little oxidation of

Mn(II) occurs in solutions below pH 8. In addition, biological oxidation of Mn(II) does

not proceed rapidly in the presence of Fe(II), and thus it is not removed significantly in

aerobic wetlands where the concentration of ferrous iron exceeds 1 mg L-1

. Manganese

does not readily form an insoluble sulfide phase and thus it is not removed to a great

extent in anaerobic wetlands, though it may precipitate in these systems as rhodocrosite

(MnCO3) (Hallberg and Johnson, 2005).

11

Permeable reactive barriers (PRBs) are being increasingly used to treat a wide range of

polluted ground waters. Some have been installed to bioremediate AMD. The

construction of PRBs involves digging of a trench or pit in the flow path of

contaminated groundwater, filling the void with reactive materials (a mixture of organic

solids and possibly limestone gravel) that are sufficiently permeable to allow free flow

of the groundwater, and landscaping of the disturbed surface. Reductive microbiological

processes within the PRB generate alkalinity (which is further enhanced by dissolution

of limestone and/or other basic minerals) and remove metals as sulfides, hydroxides,

and carbonates (Johnson and Hallberg, 2005).

As manganese is generally not efficiently removed by current passive technologies

applied to remediate mine water, alternative approaches are needed (Hallberg and

Johnson, 2005).

Another possibility to remove manganese includes their adsorption onto active sites

such as those found on algae and/or organic and inorganic compounds (Soto et al.,

2005). The use of adsorbent/ion exchange resin is a promising alternative method for

the removal of metals in dilute solution (Laus et al., 2005).

3.3.1. Manganese removal by adsorption/ion exchange

Adsorption processes may be used to remove metals, as soluble ions or complexes, from

water solutions. It represents an interesting alternative to be used when it is desired to

get very low residual concentrations of contaminants in water (Soto et al., 2005). In the

case of manganese removal by adsorption/ion exchange, just a few works have been

reported (Korn et al., 2004; Soto et al., 2005; Bosco et al., 2005, Mohan and Chander,

2006b).

Korn et al. (2004) evaluated the separation and concentration of trace manganese,

copper and zinc from saline waters using adsorption of these metal ions onto the resins

Dowex 1X8-50 and Amberlite XAD-7 in the range of pH between 3.0 and 10.0. It was

observed that a maximum amount of Mn(II) removed was from the saline water at pH

12

9.0 for Amberlite XAD-7 and pH 8.0 for resin Dowex 1X8-50. The maximum loads

obtained were 67 μg g-1

Dowex 1X8-50 and 1000 μg g-1

Amberlite XAD-7.

The removal of manganese present in aqueous solutions of concentration below 40 mg

L-1

was investigated by Soto et al. (2005) using palygorskite (philossilicate) as an

adsorbent in the range of pH between 4.5 and 8.5. The maximum loading of manganese

(6.28 mg g-1

) occurred at pH 7.5 since, at pH 8.5, manganese precipitated, possibly as

some kind of hydroxide.

Bosco et al. (2005) conducted studies on the ion exchange scolecite, a natural zeolite

from Brazil, applied in the purification of water contaminated with heavy metals. The

maximum retention of these metals occurred at pH around 6, and the decreasing order

of adsorption of metals was Cr>Mn>Cd>Ni. The maximum retention of manganese was

75%.

Mohan and Chander (2006a) used lignite for the remediation of ferrous, ferric, and

manganese ions in single, binary, ternary, and multicomponent systems from

water/wastewaters. They showed that the sorption capacity increased with the increase

in temperature in the case of Fe(II). The increase in temperature from 10 to 40ºC

doubled the adsorption capacity from 24 to 46 mg g-1

. On the other hand, Mn(II)

adsorption decreased with the increase in temperature. A decrease in temperature from

40 to 25ºC increased sorption capacity from 7.7 to 25 mg g-1

. The maximum adsorption

capacities of the lignite were 46.46 (40ºC), 28.11 (10ºC), and 11.90 (25ºC) mg g-1

for

Fe(II), Mn(II), and Fe(III), respectively. They also considered that an ion exchange

process must be the predominant mechanism in the sorption of metals on lignite.

Aziz and Smith (1992) reported the removal of manganese from water using limestone,

crushed brick and gravel. At a final pH value of 8.5, limestone removed 95% of

manganese, crushed brick 82%, gravel about 60% and the removal for aeration and

settlement with no solid media was less than 15%. But the results indicate that the

presence of carbonates must have precipitated the manganese in water, which is not

interesting for adsorption. It forces the system to control pH to prevent this unwanted

precipitation.

13

Aziz and Smith (1996), in another study, developed a filtration technique using crushed

dolomite to remove the manganese (1 mg L-1

) present in synthetic solutions at pH 7.

According to these authors, dolomite was the most effective material for the removal of

manganese. To optimize the use of dolomite as a filtering material, tests were conducted

where the best removal occurred for particles with size of 500 mm and a flow rate of 10

mL min-1

. The authors suggest that the dolomite, having a low price and good

adsorption of manganese, should be employed for water filtration systems.

Silva et al. (2010) evaluated limestone as an adsorbent for Mn removal. Batch

experiments were carried out with synthetic solutions, at initial pH 5.5 and 8.3 g

limestone L-1

and similarly, continuous tests were performed with 16.5 mg L-1

Mn2+

mine water, at initial pH 8.0 and 20.8 g limestone L-1

. Best results were achieved with

calcite limestone and its fine grinding proved to be the most effective parameter for

manganese removal. In either synthetic solutions or industrial effluents, the final

manganese concentration was below 1 mg L-1

. A neutralization step to increase the

effluent pH before efficient metal removal was necessary.

Robinson-Lora and Brennan (2010a) reported the role of chitin and chitin-associated

proteins in the removal of manganese from AMD under abiotic and anaerobic

conditions. The results showed that both types of biomolecules are capable to sorb

manganese, however, the removal capacity of the chitinous materials significantly

increases when chitin-associated proteins are present. At higher pH regimes, qm values

ranged from 0.165 (at pH 5.4) to 0.981 mg g−1

(at p . ) for “pure” chitin and

increased from 0.878 (at pH 5.2) to 5.437 mg g−1

(at pH 8.6) when both chitin and

protein were present. Results clearly suggest that the chitin-associated proteins offer

additional sorption sites for manganese.

Chitosan microspheres were shown by Laus et al. (2005) to be a promising material for

remediation of acidity, iron and manganese removal from water contaminated by

mining operation. Around 0.4 g of chitosan microspheres was necessary to remove

nearly 90% of manganese (II). The results regarding the removal of manganese (II) are

14

interesting, since there are few alternatives for water recovery with this type of

contamination.

3.4. Bone char

Bone char has been extensively used as an adsorbent for the decolorization of cane and

beet sugar and to a lesser extent for the defluoridation of drinking water. After use, it is

regenerated by washing and calcining so that the char can pass through many operating

cycles before its activity has decreased to an unacceptably low level (Choy et al., 2004).

Results for the removal of metal ions from wastewaters (Choy et al., 2004; Guedes et

al., 2007; Pan et al., 2009; Ozawa et al., 2003; ’Connor and Weatherley, 199 ) have

been also reported.

Bone char is produced by the pyrolysis of crushed animal bones. After being ground to

the appropriate particle size, the bone fragments are calcined at 800oC under controlled

conditions. X-ray diffraction reveals that bone char is a mixed adsorbent composed

approximately of tricalcium phosphate (70-76 wt%), carbon content (9-11 wt%) and

calcium carbonate (7-9 wt%) (Guedes et al., 2007). Structurally, calcium phosphate is

in the hydroxyapatite form. The amorphous carbon fraction is distributed throughout the

whole of the hydroxyapatite structure, but mostly exists as a highly active thin film over

about 50% of the porous of hydroxyapatite surface (Choy et al., 2004).

Although production and use of bone char in large scale has been currently done in

many countries, mainly for industrial sugar decolorization, the scientific literature on

the preparation and use of bone char as an adsorbent is very small compared to the

amount of literature on the other types of sorbents (Oliveira and Franca, 2008).

Scientific publications on the application of bone char as adsorbent has aroused interest

(Cheung et al., 2000; Cheung et al., 2001; Ko et al., 2000; Ko et al., 2004; Choy et al.,

2004; Choy and McKay, 2005a,b; Ribeiro, 2011).

The sorption of metal ions copper, cadmium, and zinc onto bone char was demonstrated

by Choy et al. (2004). Equilibrium studies have been analyzed using the Langmuir

isotherm equation and the maximum sorption capacities for the metals were 45.1, 53.6,

15

and 33.0 mg g−1

bone char for copper, cadmium and zinc ions, respectively. Guedes et

al. (2007) also studied the sorption of metal ions onto bone char and the maximum

sorption capacities for the metals were 25.7, 27.0 and 25.3 mg g−1

bone char for copper,

cadmium and zinc ions, respectively.

Pan et al. (2009) reported that swine bone char is effective in removing cobalt from

solution. Batch kinetics studies showed that a rapid uptake occurred within the first 5

min. When the ratio of cobalt to bone char was greater than 0.3, nearby 1.0 mol of

calcium was released as 1.0 mol of cobalt was adsorbed. The measurements of calcium

concentrations in the solution containing the metal ions suggested that ion-exchange is

the most significant mechanism for the removal of cobalt from the solution.

Hassan et al. (2008) used camel bone char for quantitative removal of Hg(II) from

wastewater. The removal of mercury from 100 mL of 10 mg L-1

of Hg(II) was achieved

by 0.03 g of camel bone charcoal at pH 2. The Langmuir adsorption capacity was 28.3

mg of Hg(II) per gram of the adsorbent.

Cheung et al. (2000) reported the sorption capacity of bone char for copper and zinc

ions (47.7 and 34.7 mg g−1

, respectively) and concluded that bone char is a suitable

sorbent for the two metal ions.

Cheung et al. (2001), in another study, evaluated the adsorption of cadmium ions onto

bone char. Equilibrium tests were done to evaluate the sorption capacity of bone char

for cadmium ions and experimental results showed it to be 64.1 mg g−1

at an

equilibrium solution concentration of 337 mg L−1

. Since the sorption capacity is

relatively high, it was concluded that bone char can be considered as a suitable sorbent

for the adsorption of cadmium in wastewater treatment systems.

The removal of As(V) from aqueous solutions using bone char was studied by Chen et

al. (2008). The adsorption was found to be strongly dependent on pH, dosage of

adsorbent, and contact time. Bone char removed 99.18% of As(V) at the initial As(V)

concentration of 0.5 mg L−1

and pH 10.

16

Ozawa et al. (2003) used fish bone hydroxyapatite for manganese removal from

laboratory solution. The initial concentration of Mn(II) was 16 mg L-1

and 0.3 g of the

bone powders was added to each 2 L of aqueous solution. After the mixture of powder

and solution to be stirred for 2 hours and held for 6 days without stirring, the maximum

adsorption capacity reached was 17 mg g-1

.

According to ’Connor and Weatherley (1997) the hydroxyapatite component of the

bone char plays an important role in the adsorption of manganese, with adsorption

occurring by ion exchange. Over 67% of manganese removals in twenty four hours for

an initial concentration of 15 mg L-1

using bone char for manganese adsorption from

potable water supplies.

Gonçalves (2006) studied the use of bone char in alkaline medium for manganese

removal. 22% of manganese present in AMD effluent using 0.2 g of bone char and 200

mL of effluent for an initial concentration of 173 mg L-1

.

3.5. Adsorption

Adsorption can be described as mass transfer phenomena of solutes (adsorbate) from a

fluid phase to a solid surface used as adsorbent. It is an operation that can solve or

attenuate the problems of pollutants dissolved in liquid effluent and also those related to

constituents in low concentrations in industrial processes, which need to be recovered

due to their high added value (Gonçalves, 2006).

Adsorption is normally classified as physical or chemical process depending on the kind

of interaction or strength existing between the adsorbent and the adsorbate. Physical

adsorption occurs when weak interparticle bonds exist between the adsorbate and the

adsorbent. Physical adsorption is easily reversible, in the majority of cases, what may

lead to the adsorbent reutilization. Chemical adsorption, on the other hand, occurs when

strong interparticle bonds are present between the adsorbate and the adsorbent.

Chemisorption is deemed to be irreversible in the majority of cases (Reynolds and

Richards, 1995).

17

3.5.1. Adsorption isotherms

The adsorption isotherm is, as well, an important tool to predict thermodynamic

parameters. Several isotherm equations are available and the most common isotherms

which are considered in many studies are the Langmuir and Freundlich isotherms.

3.5.1.1. Langmuir isotherm equation

The Langmuir adsorption isotherm, based on a theoretical model (Gibbs equation),

assumes monolayer adsorption over an energetically and structurally homogeneous

adsorbent surface. The saturation monolayer can be represented by the expression:

e

em

ebC1

bCqq

(3.5)

and a linear form:

emmeCbq

1

q

1

q

1

(3.6)

in which qe (mg g−1

) is the adsorption capacity by weight at equilibrium, qm (mg g−1

) is

the theoretical maximum adsorption capacity by weight, and b (L mg−1

) represents the

Langmuir constant, while Ce (mg L−1

) is the concentration of adsorbate at equilibrium

(Annadurai et al., 2008).

3.5.1.2. Freundlich isotherm equation

Freundlich proposed an empirical isotherm equation given by:

n

eFe CKq/1

(3.7)

or the linear form:

Fee

KlnClnn

1qln (3.8)

18

in which KF (mg1-(1/n)

L1/n

g−1

) represents the relative adsorption capacity and n is

related to the intensity of adsorption, with values of n>1 indicating favorable

adsorption. As the Freundlich isotherm equation is exponential, it can only be

reasonably applied in the low to intermediate concentration ranges (Annadurai et al.,

2008).

3.5.2. Adsorption kinetics

In adsorption systems, the mass transfer of solute or sorbate onto and within the sorbent

particle directly affects the adsorption kinetics. In general, adsorption theory is based on

the principle that four steps are involved in the process, any of which could be the rate-

controlling step. The four steps are (Levenspiel, 2003):

Transport of sorbate from the bulk of the solution to the boundary layer;

Movement of sorbate across the external liquid film boundary layer to external

surface sites;

Migration of sorbate within the pores of the sorbent by intraparticle diffusion;

Sorption of sorbate at internal surface sites.

There are, in the literature, several models proposed to fit the experimental data

obtained by adsorption process. Among these, some are based on adsorption

capacity of adsorbent material, which treats the data without considering, none of

the steps involved in the adsorption process; the most used of these are the pseudo-first

order and the pseudo-second order. However, there are also those that allow

analyzing the mechanisms involved in the adsorption process, which may be related to

the occurrence of a chemical reaction and/or mechanisms diffusion (Crini and Badot,

2008). In these cases, the most used are the diffusion in the film and intraparticle

diffusion. In order to examine the mechanism controlling the adsorption processes, such

as mass transfer and chemical reactions, several kinetic models are used to fit

experimental data.

3.5.2.1. Pseudo-first-order equation

The pseudo-first-order kinetic equation or the so-called Lagergren equation has the

following formulation:

19

te1

t qqkdt

dq

(3.9)

After definite integration by applying the initial conditions qt = 0 at t = 0 and qt = qt at t

= t, equation (3.9) becomes:

t303.2

kqlogqqlog 1

ete (3.10)

in which qt is the amount of adsorbate adsorbed at time t, qe is its value at equilibrium

and k1 (min-1

) is a constant of the pseudo-first-order adsorption. The equilibrium

adsorption capacity qe and the first-order constants k1 can be determined experimentally

from the slope and the intercept of plot of log(qe - qt) versus t. Generally, the pseudo-

first-order equation fits the data when the rate-controlling step is the diffusion across the

external liquid film boundary layer (Lagergren, 1898, apud Annadurai et al., 2008).

3.4.2.2. Pseudo-second-order equation

The pseudo-second-order equation based on adsorption equilibrium capacity may be

expressed in the form:

2

te2

t qqkdt

dq

(3.11)

Integrating equation (3.12) and applying the initial conditions described above:

tq

1

qk

1

q

t

e

2

e2t

(3.12)

in which k2 (g mg-1

min-1

) is a constant of the pseudo-second-order adsorption. The

equilibrium adsorption capacity qe and the second-order constants k2, can be determined

experimentally from the slope and the intercept of plot t/qt versus t. In general, the

pseudo-second-order expression is used to describe chemisorption (Ho, 2006).

20

3.5.2.3. Intraparticle-diffusion model

Intraparticle diffusion has a significant role in many adsorption processes and assumes

that diffusion of adsorbate occurs within the pore structure of the adsorbent. The

intraparticle diffusion model can be described as:

CtKq pt 2/1. (3.13)

where kp (mg g-1

min-0.5

) is the intraparticle diffusion constant, evaluated as the slope of

the curve qt versus t1/2

and C (mg g-1

) is the constant related to the boundary layer

thickness. If the regression of qt versus t1/2

is linear and passes through the origin, then

intraparticle diffusion is the sole rate-limiting step. However, if this line does not

pass through the origin, this indicates some degree of control associated with

the diffusion boundary layer. In this case, it can occur due to the fact that either (i)

intraparticle diffusion is not the rate-limiting step of the process, or that (ii) beyond

this step, another step in the adsorption mechanism also requires a time significant to

occur, so together controlling the rate of adsorption process (Weber and Morris, 1963

apud Annadurai et al., 2008).

3.5.3. Fixed bed adsorption

Batch adsorption experiments are easily used in laboratory scale for the treatment of

small volume of effluents, but it is less convenient to use on industrial scale, where

large volumes of wastewater are continuously generated. Batch adsorption provides

certain preliminary information such as the pH for maximum adsorption, maximum

initial metal ion concentration, and particle size for optimum adsorption of metal ions,

and approximate time for adsorption of metal ion as well as the adsorption capacity of

the adsorbent. All these information are useful for packed bed studies.

In fixed bed adsorption, the solute concentration in the solid is initially zero. Feed

solution is added into the fixed bed, giving the abrupt concentration profiles that move

through the bed. The concentration in the solid is either zero or in equilibrium with the

concentrated feed. In the stirred tanks, the situation is very different. Both solid and

solution are stirred together. At very short time, the solution concentration is higher than

21

that in the solid. After a while, mass transfer increases the concentration in the solid and

depletes that in solution. At long time, the solid reaches an equilibrium, but with the

depleted solution. This equilibrium concentration is much less than that with the feed

solution. Thus the stirred tank has a much less effective separation than the fixed bed

(Cussler, 1997).

In fixed bed adsorption, the concentration in the liquid phase and in the solid phase

change with time as well as with position in the bed. At first, most of the mass transfer

takes place near the inlet of the bed, where the fluid first contact the adsorbent. With

time, the mass transfer zone moves down the bed. Although the packed bed systems can

be fed with flows upward or downward, it is important to note that flows upward under

high flow may fluidize the bed, causing friction between the particles. In excess, this

can result in reduction of average particle size of the bed and therefore the loss of the

material adsorbent (McCabe et al., 2005).

The design and theory of fixed bed adsorption systems focuses on establishing the shape

of the breakthrough curve and its velocity through the bed. Breakthrough curves versus

time plots can be easily plotted because this is the way in which data are most

commonly attained. However, it can sometimes gain more insight into the behavior of

these beds by plotting breakthrough curves versus the treated volume or versus the

number of bed volumes. Plotting a breakthrough curve versus treated volume is useful

to compare different adsorbents. For a nonadsorbing solute, the breakthrough occurs

when the eluted volume equals the void volume in the fixed bed. For a strongly

adsorbing solute, the breakthrough occurs when the treated volume is much greater than

the void volume in the fixed bed. As a result, the adsorption data are not plotted versus

time or eluted volume, but versus the number of bed volumes, that is the eluted volume

divided by the volume of bed’s voids (Cussler, 1997).

Measuring the concentration of adsorbate in the stream leaving the column and

constructing a graph with these values in function of number of bed volumes or time,

results in the breakthrough curve as shown in Figure 3.2 (McCabe et al., 2005).

22

Figure 3.2: Breakthrough curve for adsorption in a fixed bed (Adapted from Reynolds

and Richards, 1995).

Figure 3.2 describes a process which presents a fixed bed column with height h, fed

with a solution of concentration C0, pumped at a continuous flow rate Q. The

breakthrough point represents the point where the adsorbate starts to be detected at the

outlet of the column, so it is normally used to determine the capacity of a given column.

3.5.3.1. Adsorption kinetics in fixed bed systems

Although the fixed bed adsorption follows the same steps of the batch adsorption

process, the mass transfer phenomena is not necessarily the same. Therefore, data

obtained in experimental systems in fixed bed adsorption are evaluated and described

by different models.

The Thomas model is commonly used in the modeling of fixed bed breakthrough curves

in environmental sorption research (Unuabonah et al., 2010). It is widely used for the

design of activated carbon adsorbers due to theirs simplicity. The Thomas model

assumes a Langmuir (favorable) isotherm (Chu, 2010).

23

The expression by Thomas for an adsorption column is given as follows:

Ct

C0

1

1 exp q0

C0t

(3.14)

The linearized form of Thomas model can be expressed as follows:

n C0

Ct

1 q0

C0 (3.15)

in which kT (L mg-1

min-1

) is the Thomas rate constant, q0 is the sorption capacity of the

adsorbent per unit mass (mg g-1

), M is the mass of adsorbent (g) and Q is the flow rate

(L min-1

).

The kinetic coefficient kT and the adsorption capacity q0 can be obtained from the plot

of ln C0

Ct

1 against time (t) at a given flow rate using linear regression (Cavas et al.,

2011).

3.6. Desorption

Adsorption studies can be complemented with desorption ones. Such studies aims to

recover the metal retained and reuse the adsorbent in subsequent loading and unloading

cycles. When the adsorbent becomes exhausted or when the effluent from the adsorbent

bed reaches the maximum allowable discharge level, the recovery of the adsorbent

material becomes necessary. Regeneration of spent adsorbent columns is quite an

important process in wastewater treatment. In this process, 10–20% of the adsorbent is

usually lost by attrition during each cycle and the recovery of the adsorbate is also not

possible (Mohan and Chander, 2006b).

24

Desorption studies of lignite loaded with manganese were performed by Mohan and

Chander (2006b). To simulate industrial conditions for acid mine wastewater treatment,

all tests were carried out using single and multi-columns setup in down flow mode.

Desorption of metal ions in three column systems was conducted with 0.1 mol L-1

HNO3 and showed that the recovery of the metal ions was almost 100%.

Desorption experiments with bone char loaded with Zn were carried out by Guedes et

al. (2007). The first desorption test was carried out with 10 g L-1

of bone char in

distilled water at pH 5.4. The second desorption experiment was carried out with

sulfuric acid solution (0.01 mol L-1

) at the same conditions as for distilled water. It was

observed that desorption with distilled water was not effective but in the case of

desorption carried out with the sulfuric acid solution, the levels of desorption was 40%.

3.7. References

ANNADURAI, G.; LING, L.Y.; LEE, J.F. Adsorption of reactive dye from solution by

chitosan: isotherm, kinetic and thermodynamic analysis. Journal of Hazardous

Materials, v. 152, p. 337-346, 2008.

AZIZ, H.A.; SMITH, P.G. Removal of manganese from water using crushed dolomite

filtration technique. Water Research, v. 30, p. 489-492, 1996.

AZIZ, H.A.; SMITH, P.G. The influence of pH and course media on manganese

precipitation from water. Water Research, v. 26, p. 853-855, 1992.

BAMFORTH, S.M.; MANNING, D.A.C.; SINGLETON, I.; YOUNGER, P.L.;

JOHNSON, K.L. Manganese removal from mine waters – investigating the occurrence

and importance of manganese carbonates. Applied Geochemistry, v. 21, p. 1274-1287,

2006.

25

BOSCO, S.M.D.; JIMENEZ, R.S.; CARVALHO, W.A. Removal of toxic metals from

wastewater by Brazilian natural scolecite. Journal of Colloid and Interface Science, v.

281, p. 424-431, 2005.

CAVAS, L.; KARABAYA, Z.; ALYURUKA, H.; DOGAN, H.; DEMIR, G.K. Thomas

and artificial neural network models for the fixed-bed adsorption of methylene blue by a

beach waste Posidonia oceanica (L.) dead leaves. Chemical Engineering Journal, v.

171, p. 557-562, 2011.

CHEN, Y.-N.; CHAI, L.-Y.; SHU, Y.-D. Study of arsenic(V) adsorption on bone char

from aqueous solution. Journal of Hazardous Materials, v. 160, p. 168-172, 2008.

CHEUNG, C.W.; PORTER, J.F.; McKAY, G. Sorption kinetic analysis for the removal

of cadmium ions from effluents using bone char, Water Research, v. 35, p. 605-612,

2001.

CHEUNG, C.W.; PORTER, J.F.; McKAY, G. Sorption kinetics for the removal of

copper and zinc from effluents using bone char. Separation and Purification

Technology, v. 19, p. 55-64, 2000.

CHOY, K.K.H.; McKAY, G. Sorption of cadmium, copper, and zinc ions onto bone

char using Crank diffusion model. Chemosphere, v. 60, p. 1141-1150, 2005a.

CHOY, K.K.H.; McKAY, G. Sorption of metal ions from aqueous solution using bone

char, Environment International, v. 31, p. 845-854, 2005b.

CHOY, K.K.H.; BARFORD, J.P.; McKAY, G. Production of activated carbon from

bamboo scaffolding waste-process design, evaluation and sensitivity analysis, Chemical

Engineering Journal, v. 109, p. 147-165, 2005.

26

CHOY, K.K.H.; KO, D.C.K.; CHEUNG, C.W.; PORTER, J.F.; McKAY, G. Film and

intraparticle mass transfer during the adsorption of metal ions onto bone char. Journal

of Colloid and Interface Science, v. 271, p. 284-295, 2004.

CHU, K.H. Fixed bed sorption: Setting the record straight on the Bohart–Adams and

Thomas models. Journal of Hazardous Materials, v. 177, p. 1006-1012, 2010.

CONSELHO NACIONAL DO MEIO AMBIENTE. Resolução CONAMA n.º

357/2005: Classificação dos corpos de água e diretrizes ambientais para o seu

enquadramento, bem como estabelece as condições e padrões de lançamento de

efluentes. Brasília, 2005. Disponível na Web em:

http://www.mma.gov.br/port/conama/index.cfm.

CRINI, G., BADOT, P. Application of chitosan, a natural aminoplysaccharide, for dye

removal from aqueous solutions by adsorption processes using batch studies: a review

of recent literature. Progress in Polymer Science, v. 33, p. 399-447, 2008.

CUSSLER, E.L. Diffusion: Mass Transfer in Fluid Systems. 2nd edition. Cambridge

University Press, p. 312-314, 1997.

GONÇALVES, C.R. Remoção de manganês e recuperação de urânio presentes em

águas ácidas de mina. Belo Horizonte, 2006. Dissertação (Mestrado) - Centro de

Desenvolvimento da Tecnologia Nuclear.

GUEDES, T.S.; MANSUR, M.B.; ROCHA, S.D.F. A perspective of bone char use in

the treatment of industrial liquid effluents containing heavy metals. In: XXI ENTMME,

Ouro Preto - MG, 2007.

HALLBERG, K.B.; JOHNSON, D.B. Biological manganese removal from acid mine

drainage in constructed wetlands and prototype reactors. Science of the Total

Environment, v. 338, p. 115-124, 2005.

27

HASSAN, S.S.M; AWWAD, N.S.; ABOTERIKA, A.H.A. Removal of mercury(II)

from wastewater using camel bone charcoal. Journal of Hazardous Materials, v. 154, p.

992-997, 2008.

HO, Y.S. Review of second-order models for adsorption systems. Journal of Hazardous

Materials, B136, p. 681-689, 2006.

JOHNSON, D.B.; HALLBERG, K.B. Acid mine drainage remediation options: a

review. Science of the Total Environment, v. 338, p. 3-14, 2005.

KO, D.C.K.; PORTER, J.F.; McKAY, G. Optimised correlations for the fixed-bed

adsorption of metal ions on bone char. Chemical Engineering Science, v. 55, p. 5819-

5829, 2000.

KO, D.C.K.; CHEUNG, C.W.; CHOY, K.K.H.; PORTER, J.F.; McKAY, G. Sorption

equilibria of metal ions on bone char, Chemosphere, v. 54, p. 273-281, 2004.

KORN, M.G.A., SANTOS JR., A.F., JAEGER, H.V., SILVA, N.M.S., COSTA, A.C.S.,

2004. Cooper, zinc and manganese determination in saline samples employing FAAS

after separation and preconcentration on Amberlite XAD-7 and Dowex 1X-8 loaded

with alizarin Red. Journal of Brazilian Chemical Society, 15(2), 212-218.

LANGERGREN. Zur theorie der sogenannten adsorption geloster stoffe, Kungliga

Svenska Ventenskapsaka demiens, Handlingar 24, p. 1–39, 1898 apud ANNADURAI,

G.; LING, L.Y.; LEE, J.F. Adsorption of reactive dye from solution by chitosan:

isotherm, kinetic and thermodynamic analysis. Journal of Hazardous Materials, v. 152,

p. 337-346, 2008.

LAUS, R.; LARANJEIRA, M.C.M.; MARTINS, A.O.; FÁVERE, V.T. Microesferas de

quitosana reticuladas com tripolifosfato utilizadas para remoção da acidez, ferro(III) e

manganês(II) de águas contaminadas pela mineração de carvão. Quimica Nova, v. 29, no

1, p. 34-39, 2006.

28

LEVENSPIEL, O. Engenharia das reações químicas. Tradução da 3ª edição. São Paulo:

Editora Blucher, 2003. p. 480-497.

LIND, C.J.; HEM, J.D. Manganese minerals and associated fine particulates in the

streambed of Pinal Creek, Arizona, USA: a mining-related acid drainage problem.

Applied Geochemistry, v. 8, p. 67-80, 1993.

LIU, Y.; LIU, Y.J. Biosorption isotherms, Kinetics and thermodynamics. Separation

and purification Technology, v. 61, p. 229-242, 2008.

McCABE, W.L.; SMITH, J.C.; HARRIOTT, P. Unit operations of chemical

engineering: Chemical Engineering Series. 7th edition. McGraw-Hill´s Science, p. 836-

847, 2005.

MOHAN, D.; CHANDER, S. Single, binary, and multicomponent sorption of iron and

manganese on lignite. Journal of Colloid and Interface Science, v. 299, p. 76-87, 2006a.

MOHAN, D.; CHANDER, S. Removal and recovery of metal ions from acid mine

drainage using lignite - A low cost sorbent. Journal of Hazardous Materials, v. B137, p.

1545-1553, 2006b.

NASCIMENTO, M.R.L. Remoção e recuperaçâo de urânio de águas ácidas de mina por

resina de troca iônica. São Carlos, 1998. Dissertação (Mestrado) - Universidade Federal

de São Carlos.

’C NN R, J. . AND WEA ERLEY, L.R. An investigation into the mechanism of

adsorption of humic substances and trace metal compounds from potable water

supplies. Global Environmental Biotechnology, p. 91-106, 1997.

OLIVEIRA, L.S.; FRANCA, A.S. Low-cost adsorbents from agri-food wastes. In: Food

Science and Technology: New Research, Nova Science Publishers, chapter 3, 2008.

29

OZAWA, M.; SATAKE, K.; SUZUKI, S. Removal of aqueous manganese using fish

bone hydroxyapatite. Journal of materials science letters, v.22, p.1363 – 1364, 2003.

PAN, X.; WANG, J.; ZHANG, D. Sorption of cobalt to bone char: Kinetics,

competitive sorption and mechanism. Desalination, v. 249, p. 609-614, 2009.

REYNOLDS, T.D.; RICHARDS, P.A. Unit operations and processes in environmental

engineering. 2ª Edição. Boston: PWS Publishing Company, p. 350-409, 1995.

RIBEIRO, M.V. Uso de Carvão de Osso Bovino na Defluoretação de Água para Uso

em Abastecimento Público. Belo Horizonte, 2011. Dissertação (Mestrado) - Universidade

Federal de Minas Gerais.

ROBINSON-LORA, M.A.; BRENNAN, R.A. Biosorption of manganese onto chitin

and associated proteins during the treatment of mine impacted water. Chemical

Engineering Journal. v. 162, p. 565-572, 2010a.

ROBINSON-LORA, M.A.; BRENNAN, R.A. Chitin complex for the remediation of

mine impacted water: geochemistry of metal removal and comparison with other

common substrates, Applied Geochemistry, v. 25, p. 336-344, 2010b.

ROBINSON-LORA, M.A.; BRENNAN, R.A. Efficient metal removal and

neutralization of acid mine drainage by crab-shell chitin under batch and continuous-

flow conditions. Bioresource Technology, v.100, p. 5063-5071, 2009.

ROSE, A.W.; MEANS, B.; SHAH, P.J. Methods for passive removal of manganese

from acid mine drainage. In: 2003 West Virginia Surface Mine Drainage Task Force

Symposium, Morgantown, WV, p. 11, 2003a.

30

ROSE, A.W.; SHAH, P.J.; MEANS, B. Case studies of limestone-bed passive systems

for manganese removal from acid mine drainage. In: National Meeting of the American

Society of Mining and Reclamation and the 9th

Billings Land Reclamation Symposium,

Billings MT, p. 1058-1078, 2003b.

SHEORAN, A.S.; SHEORAN, V. Heavy metal removal mechanism of acid mine

drainage in wetlands: A critical review. Minerals Engineering, v.19, p. 105–116, 2006.

SILVA, A.M.; CRUZ, F.L.S.; LIMA, R.M.F.; TEIXEIRA, M.C.; LEÃO, V.A.

Manganese and limestone interactions during mine water treatment. Journal of

Hazardous Materials, v. 181, p. 514-520, 2010.

SOTO, O.A.J.; TOREM, M.L.; TRINDADE, R.B.E. Palygorskite as a sorbent in the

removal of manganese(II) from water mine effluents. In: XIII INTERNATIONAL

CONFERENCE ON HEAVY METALS IN THE ENVIRONMENT, Rio de Janeiro -

RJ, 2005.

UNUABONAH, E.I.; OLU-OWOLABI, B.I.; FASUYI, E.I.; ADEBOWALE, K.O.

Modeling of fixed-bed column studies for the adsorption of cadmium onto novel

polymer–clay composite adsorbent. Journal of Hazardous Materials, v. 179, p. 415-

423, 2010.

VAIL, W.J.; RILEY, R.K. The Pyrolusite Process: A bioremediation process for the

abatement of acid mine drainage. Green Lands, v. 30 (4), p. 40-46, 2000 apud ROSE,

A.W.; SHAH, P.J.; MEANS, B. Case studies of limestone-bed passive systems for

manganese removal from acid mine drainage. In: National Meeting of the American

Society of Mining and Reclamation and the 9th

Billings Land Reclamation Symposium,

Billings MT, p. 1058-1078, 2003b.

WEBER, W.J.; MORRIS, J.C. Kinetic of adsorption carbon solution, Journal Sanitary

Engineering Divisor, American Society Civil Engineers v. 89, p. 31-59, 1963 apud

ANNADURAI, G.; LING, L.Y.; LEE, J.F. Adsorption of reactive dye from solution by

31

chitosan: isotherm, kinetic and thermodynamic analysis. Journal of Hazardous

Materials, v. 152, p. 337-346, 2008.

32

4. BATCH REMOVAL OF MANGANESE FROM ACID MINE DRAINAGE

USING BONE CHAR

Dalila Chaves Sicupiraa, Thiago Tolentino Silva

a, Versiane Albis Leão

b and Marcelo

Borges Mansura,*

a Departamento de Engenharia Metalúrgica e de Materiais, Universidade Federal de

Minas Gerais

Av. Antônio Carlos, 6627, Campus Pampulha, 31270-901, Belo Horizonte, MG, Brazil

Tel.: +55 (31) 3409-1811; Fax: +55 (31) 3409-1716

E-mail: marcelo.mansur@demet.ufmg.br

* Corresponding author

b Departamento de Engenharia Metalúrgica e de Materiais, Universidade Federal de

Ouro Preto

Campus Universitário, Morro do Cruzeiro, 35400-000, Ouro Preto, MG, Brasil

Abstract

Batch kinetics and equilibrium of manganese removal from acid mine drainage (AMD)

have been investigated in this study. Equilibrium tests revealed that Langmuir based

maximum manganese uptake capacity was 22 mg g-1

for AMD effluent and 20 mg g-1

for laboratory solution. Manganese kinetics onto bone char was best described by a

pseudo-second order model. Manganese removal was mainly influenced by operating

variables solid/liquid ratio and pH of the aqueous phase. In fact, manganese removal

was favored at nearly neutral pH values. The effect of particle size and temperature was

found to be not significant for the operating range investigated. The mechanism for the

removal of manganese with bone char was evaluated. It has been found that intraparticle

diffusion is the main rate-limiting step for manganese sorption systems but additional

contribution from boundary layer diffusion might also affect removal when bone char

particles of smaller sizes are used. The final concentration of fluoride and others metals

present in the AMD effluent was in agreement with the limit concentration established

by Brazilian legislation (CONAMA, 2005). The study revealed that bone char is a

suitable material to be used for the removal of manganese from AMD effluents.

Key-words: manganese; bone char; acid mine drainage; adsorption.

33

4.1. Introduction

Given its dynamics and persistence, acid mine drainage (AMD) is one of the most

serious environmental problems of the mining industry. Such phenomenon occurs when

pyrite and/or other sulfide minerals are oxidized due to its exposure to oxygen and

water, thus producing sulfuric acid that might dissolve metals species resulting in

contamination of the environment (Nascimento, 1998). The main sources of AMD are

open or underground pit mine, waste rock piles, tailings storage and ore stockpiles. The

raise in the concentration of metal ions in the water due to AMD is an important source

of contamination of watercourses, especially when ions can be spread to the food chain.

Aiming to minimize the severity of such impacts, environmental protection laws require

appropriate plans for mine closure as an attempt to increase the cycle life of closed

mines including steps of decommissioning and recovery of degraded areas (Gonçalves,

2006).

In Brazil, AMD has been highlighted in coalfields in the south of the country and also in

the Industrial Mining Complex of Poços de Caldas (Indústrias Nucleares do Brasil –

INB), in the state of Minas Gerais. AMD generated in this last region contains

radionuclides (uranium, thorium, radium, and others) and species like manganese, zinc,

fluoride and iron at levels of concentration above those allowed by Brazilian legislation

for direct discharge (Gonçalves, 2006; CONAMA, 2005). The current treatment of such

acid water consists of metals precipitation with lime followed by pH correction. Most of

metal species are precipitated but manganese ions removal from AMD is notoriously

difficult to occur due to their complex chemistry in aqueous systems (Bamforth et al.,

2006; Robinson-Lora and Brennan, 2010). For complete precipitation of manganese, the

pH must be raised to around 11; and it involves a significant consumption of lime,

around 350 tons per month (Gonçalves, 2006). In addition, after manganese removal,

the pH must be neutralized for discharge so such treatment is costly, generates large

volumes of sludges and requires the consumption of huge quantities of reagents to be

effective. Therefore, new technologies to treat AMD effluents containing manganese

are urged.

The adsorption of metal species has been investigated using various alternative

materials (Korn et al., 2004; Bosco et al., 2005; Soto et al., 2005) but relatively little

34

attention has been directed to the manganese removal from AMD solutions. Adsorption

has proven to be a particularly interesting technique in hydrometallurgical applications.

It allows the recovery of ions from very dilute solutions and it has also the ability to

process large volumes of solutions when other separating methods seem unfavorable.

Bone char has been successfully applied for the removal of various metal species such

as copper, zinc, cadmium, arsenic, mercury, etc (Cheung et al., 2000; Choy et al., 2004;

Chen et al., 2008; Hassan et al., 2008; Pan et al., 2009), however the use of bone char in

the manganese removal has little information until date ( ’Connor and Weatherley,

1997, Ozawa et al., 2003 and Gonçalves, 2006). As bone char consists basically of

hydroxyapatite (Ca10(PO4)6(OH)2) and calcite (CaCO3), its use could not only remove

manganese but also raise the pH of the effluent due to the dissolution of the calcite, thus

contributing to reduce the lime consumption currently used in the treatment of AMD

solutions containing manganese.

In the view of the aforementioned, the aim of this work is to evaluate the feasibility of

using bone char in the treatment of AMD effluents containing manganese. Batch

equilibrium and kinetics tests were carried out using AMD effluents and laboratory

solutions containing manganese at typical concentrations for comparative analysis.

4.2. Experimental

4.2.1. Reagents and instrumentation

AMD effluent generated by the industrial mining complex of Poços de Caldas

(Indústrias Nucleares do Brasil – INB) and bone char supplied by Bone Char do Brasil

Ltda were used in the present study. A laboratory solution containing 100 mg L-1

of

manganese was also prepared using chemicals of analytical reagent grade dissolved in

distilled water.

A mechanical shaker (Nova Ética 109) was used for agitating the sample solutions with

bone char. Atomic emission spectrometer with inductively coupled plasma (ICP-AES

Perkin Elmer OPTIMA 7300DV) was used for assessing the concentration of

35

manganese in the aqueous solutions. Sulfate and fluoride analysis were performed using

a DX500 Dionex ion chromatograph. The pH measurements were carried out using a

pH meter (Quimis). A scanning electron microscopy (SEM, JEOL JSM 5410) with X-

ray energy dispersive spectrometry (EDS, NORAN VOYAGER 3.4.1) was employed to

analyze the surface morphology and to obtain the elemental composition of the bone

char. The solid was also analyzed by XRD (X-Ray Diffraction, using a PANalytical

X’Pert APD diffractometer) and F IR (Fourier ransform Infra-Red, using a Perkin

Elmer Paragon 1000 spectrometer). In addition, surface area, total pore volume and

average pore diameter were assessed by measuring multipoint N2 isotherms in a BET

Quantachrome model Autosorb 1C.

4.2.2. Manganese removal studies

4.2.2.1. Batch sorption kinetics

Kinetics tests were conducted with 400 mL of laboratory solution or effluent at room

temperature and samples of aqueous solutions was collected with time. The solutions

were placed with 2 g of bone char (solid/liquid ratio = 2/400 g mL-1

) and the mixture

was magnetically stirred for 48 hours at a rotation speed of 150 min-1

, keeping constant

pH due to the buffer effect from the calcite dissolution. Samples of the aqueous

solutions (1 mL) were collected at fixed contact times, diluted, filtered and acidified

using HNO3 before being analyzed by ICP-AES. The influence of the solid/liquid ratio

on the sorption kinetics of manganese was studied using different amounts of bone char,

i.e., 1 and 3 g (solid/liquid ratio 1/400 and 3/400 g mL-1

, respectively). Kinetics sorption

studies were also carried out using bone char samples at different particle size (417-833

µm, 104-147 µm and <53 µm) to determine the effect of particle sizes.

4.2.2.2. Batch sorption isotherms

In these experiments, a series of dilutions were prepared at concentration from 10 to 100

mg L-1

from the original effluent and laboratory solution. Aliquots of 100 mL of each

dilution were placed into erlenmeyer flasks. The pH of effluent was first adjusted to 5.0

or 6.5 using Ca(OH)2 solution (30%) in order to precipitate other metals and filtered. No

pH adjustment was done to the laboratory solutions (initial pH = 5.7). In each flask,

36

250mg of bone char was added and the mixtures were shaken at 150 rpm and room

temperature (25ºC) using a mechanical shaker for 48 hours. The effect of the following

operating variables was investigated: initial pH of effluent (5.0 and 6.5) and temperature

(10, 25 and 40ºC). After agitation, the pulp was filtered and the filtrate analyzed by ICP-

AES to assess the metal concentration.

4.3. Results and discussion

4.3.1. Characterization analysis

4.3.1.1. Characterization of the bone char

The bone char has a real density of 2.9 g cm-3

. Its pore diameter includes mesopores.

Table 1 shows the pore distribution for the particle size range studied. As a consequence

of its highly porous nature (total pore volume 0.275 cm3

g-1

), bone char shows a relative

high surface area of 93 m2

g-1

.

Table IV.1: Pore distribution for the particle size range studied.

Pore sizes (Å) Pore distribution for 417-

833 µm (%)

Pore distribution for <53

µm (%)

< 20 0.01 0.00

20-50 13.49 5.28

50-80 21.52 16.96

80-110 21.41 15.87

110-140 21.12 12.35

140-170 7.37 8.66

170-200 4.70 12.28

200-400 10.38 28.61

On the regard of composition, bone char contains both organic and inorganic

compounds. According to the XRD analysis of the bone char (Figure 4.1), the main

37

phases presents are hydroxyapatite (Ca10(PO4)6(OH)2) and calcite (CaCO3). The

presence of such species is corroborated by FTIR spectrum analysis. Their characteristic

vibrational bands are shown in Figure 4.2 (hydroxyl group at 3420 and 1630 cm-1

,

phosphate group at 1035, 603 and 565 cm-1

and the carbonate group at 1457 cm-1

).

Figure 1.1: X-ray pattern of bone char.

Figure 4.2: FTIR spectra of bone char using KBr discs

The morphology and the chemical composition of the bone char assessed by SEM-EDS

before the contact with the AMD effluent, figures 4.3a and 4.3b, show a significant

38

presence of calcium and phosphorus, as expected. After the adsorption process (Figure

4.3c and 4.3d), a decrease in the relative Ca/P ratio was observed due to the dissolution

of calcite. It was verified also the presence of manganese and others species (F, S, La,

Ce), thus suggesting that such elements are present on the surface of the loaded bone

char possibly due to some adsorptive and/or precipitating process.

Figure 4.3: Morphology and chemical composition of bone char by SEM-EDS: (a, b)

bone char as received, and (c, d) bone char after contact with AMD effluent.

4.3.1.2. Characterization of the AMD effluent

The AMD effluent used in this study was collected nearby the uranium mine of INB

which is located in Poços de Caldas, Brazil. The composition of the effluent before and

after precipitation with Ca(OH)2 and the limit of resolution CONAMA 357 (Brazilian

standard for effluent discharge) is shown in Table IV.2. It can be seen that manganese

concentration in the AMD exceeds the maximum value of the Brazilian legislation. The

same occurs to zinc and acidity, so treatment of effluent is mandatory. On the other

(a) (b)

(c) (d)

39

hand, the concentration of iron was found to be acceptable for discharge. After the

precipitation with Ca(OH)2, it can be seen that manganese concentration is still not in

agreement with the Brazilian legislation, but the concentration of the others elements

present in the AMD effluent are in agreement with it, except fluoride.

Table IV.2: Chemical characterization of the AMD effluent.

Parameters Concentration

(mg L-1

)

Concentration after

precipitation

(mg L-1

)

CONAMA 357

(mg L-1

)

U 6.8 <0.3 -

Mn 107.5 89.5 1.0

Ca 104.9 428 -

Mg 7.6 7.4 -

Al 164.2 28.5 -

Zn 17.7 4.0 5.0

Fe <0.01 <0.01 15.0

F- 99.0 38.0 10.0

SO42-

1349 1335 -

pH* 2.97 5.64 6 to 9

* All parameters are expressed in mg L-1

, except pH.

- Permissible level not defined by the Brazilian legislation.

4.3.2. Batch sorption kinetics

The kinetic equation of pseudo-first order was found to be not adequate to fit the

kinetics data for manganese removal (R2 ≤ 0. 1 for all cases); in fact, a non-linear

behavior over the time was obtained indicating that more than one mechanism is

involved in the adsorption process. On the other hand, the kinetic equation of pseudo-

second order fitted experimental data satisfactorily (R2 ≥ 0.9 for all cases). The

linearized pseudo-second order equation is expressed as:

40

e

2

e2t q

t

qk

1

q

t

(4.1)

in which qt (mg g-1

) is the amount of manganese adsorbed in the bone char at time t, qe

(mg g-1

) is its value at equilibrium and k2 (g mg-1

min-1

) is a constant of the pseudo-

second-order adsorption model. The equilibrium adsorption capacity qe and the second-

order constants k2 were determined experimentally from the slope and the intercept of

plot t/qt versus t, respectively. In general, the pseudo-second order model is used to

describe chemisorption mechanisms (Ho, 2006).

The effect of solid/liquid ratio on the adsorption kinetics of manganese is shown in

Figure 4.4a and 4.4b for the laboratory solution containing only manganese and for the

AMD effluent, respectively. As expected, a higher percentage of manganese removal

was obtained with the increase on the solid/liquid ratio for both solutions. Also, the time

required to load the adsorbent increased with the increase on the solid/liquid ratio. The

effect of competing species present in the AMD effluent for the adsorbent is shown in

Figure 4.4b, resulting in a significant reduction on the manganese removal when

compared with laboratory solution containing only manganese. Such result points out

the necessity to raise the pH of the effluent before treatment in order to increase the

loading of the bone char for the specific removal of manganese. The effect of competing

species was verified also in the calculated qe shown in Table IV.3. In fact, smaller

values were obtained when the AMD effluent was used, as well as with the increase on

the solid/liquid ratio for both solutions as expected. The final pH values of the solutions

were kept constant in the range of 7.2 to 7.4 due to the buffer effect from the calcite

dissolution.

41

Figure 4.4: Effect of solid/liquid ratio on the kinetics of manganese removal with bone

char from (a) laboratory solution (C0 = 100 mg L-1

, pHi = 5.76), and (b) AMD effluent

(25°C, 150 min-1

, pHi = 2.96, 417-833 µm, continuous curves are pseudo-second order

model).

0,0

0,2

0,4

0,6

0,8

1,0

0 600 1200 1800 2400 3000

C/C

0

Time (min)

1/400 g/mL

2/400 g/mL

3/400 g/mL

0,2

0,4

0,6

0,8

1,0

0 600 1200 1800 2400 3000

C/C

0

Time (min)

1/400 g/mL

2/400 g/mL

3/400 g/mL

(a)

(b)

42

Table IV.3: Parameters of pseudo-second order model for the effect of solid/liquid ratio

(25°C, 150 min-1

, 417-833 µm, pHi = 5.5-5.7, C0 = 100 mg L-1

in the laboratory

solution).

Solid/liquid ratio

(g mL-1

)

qe

(mg g-1

)

k2 x 103

(mg g-1

min-1

) R

2

Laboratory

solution

1/400 19 0.33 0.99

2/400

3/400

17

13

0.48

1.00

0.99

0.99

AMD

effluent

1/400 14 1.30 0.99

2/400

3/400

10

7

0.44

1.20

0.98

0.99

Regarding the effect of particle size, the results obtained for laboratory solution (Figure

4.5a) and AMD effluent (Figure 4.5b) revealed that the different particle size ranges

studied have little effect in the manganese adsorption at longer times but loading

kinetics is affected as expected. As expected, the higher k2 values for smaller particle

size indicated that the time to reach the equilibrium is shorter for the smaller particle

size. Once again the effect of competing species present in the AMD effluent for the

adsorbent is shown in Figure 4.5b, resulting in a significant reduction on the manganese

removal. The values of qe obtained for different particle sizes are close, as indicated in

Table IV.4. The final pH of the solution was kept constant in the range of 7.2 to 7.5.

Ozawa et al. (2003) used fish bone hydroxyapatite for manganese removal from

laboratory solution. The initial concentration of Mn(II) was 16 mg L-1

and 0.3 g of the

bone powders was added to each 2 L of aqueous solution. After the mixture of powder

and solution to be stirred for 2 hours and held for 6 days without stirring, the maximum

adsorption capacity reached was 17 mg g-1

. This value of maximum loading is close to

the values obtained in this kinetic study for manganese removal using bone char.

43

Figure 4.5: Effect of particle size on the kinetics of manganese removal with bone char

from (a) laboratory solution (C0 = 100 mg L-1

, pHi = 5.76), and (b) AMD effluent (25°C,

150 min-1

, pHi = 2.96, 2/400 g mL-1

, continuous curves are pseudo-second order model).

0,0

0,2

0,4

0,6

0,8

1,0

0 600 1200 1800 2400 3000

C/C

0

Time (min)

417-833 µm

104-147 µm

<53 µm

0,2

0,4

0,6

0,8

1,0

0 600 1200 1800 2400 3000

C/C

0

Time (min)

417-833 µm

104-147 µm

<53 µm

(a)

(b)

44

Table IV.4: Parameters of pseudo-second order model for the effect of particle size

(25°C, 150 min-1

, 2/400 g mL-1

, pHi = 5.5-5.7, C0 = 100 mg L-1

in the laboratory

solution).

Particle size

(µm)

qe

(mg g-1

)

k2 x 103

(mg g-1

min-1

) R

2

Laboratory

solution

417-833 17 0.51 0.99

104-147 18 3.80 0.99

<53 18 5.60 0.99

AMD

effluent

417-833 10 0.44 0.98

104-147 10 1.60 0.99

<53 13 2.00 0.99

4.3.3. Batch sorption isotherms

The Langmuir adsorption isotherm which is based on a theoretical model (Gibbs

equation) was adequate to fit experimental data of manganese adsorption onto bone char

for the conditions investigated. The saturation monolayer can be represented by the

following linearized expression:

emme Cbq

1

q

1

q

1

(4.2)

in which qe (mg g−1

) is the adsorption capacity by weight at equilibrium, qm (mg g−1

) is

the theoretical maximum adsorption capacity by weight, b (L mg−1

) represents the

Langmuir constant, and Ce (mg L−1

) is the concentration of adsorbate in the aqueous

phase at equilibrium (Annadurai et al., 2008).

The effect of initial pH of the AMD effluent on the removal of manganese is shown in

Figure 4.6. As indicated by the equilibrium data, it is observed that the removal of

manganese is more effective when the initial pH of the effluent was raised to 6.5 due to

precipitation of some metals present in the liquor. Preliminary tests has pointed out that

pH range 7.0-7.5 would be ideal for the adsorption of manganese. However, when the

45

initial pH of effluent was set to 5.0 the final pH was 6.0, and when the initial pH was

6.5 its final value was 7.4, thus attending CONAMA 357 (see Table IV.2) in both

situations. The raise on pH during contact of AMD effluent with bone char is attributed

to the dissolution of calcite (CaCO3) present in the adsorbent. Therefore, for the

solid/liquid ratio used in this experiment, initial pH = 6.5 has a higher value of qm for

the manganese adsorption, as shown in Table IV.5. For laboratory solution, the effect of

the initial pH was not studied because it was not necessary to adjust it.

Figure 4.6: Effect of initial pH on the isotherm of manganese removal from AMD

effluent with bone char (25°C, 150 min-1

, 53 µm, 0.25/100 g mL-1

, 48 hours; continuous

lines are Langmuir model).

Table IV.5: Parameters of Langmuir equation for the pH effect for AMD effluent (25°C,

150 min-1

, <53 µm, 0.25/100 g mL-1

, 48 hours).

pHi

pHf

qm

(mg g-1

)

b x 10

(L mol-1

)

R2

5.0 6.0 14 0.43 0.98

6.5 7.4 22 0.47 0.99

0

4

8

12

16

20

0 10 20 30 40 50

q (

mg g

-1)

Ceq (mg L-1)

pHi=6.5

pHi=5.0

46

The effect of temperature on the manganese removal from laboratory solution and from

AMD effluent shown in Figure 4.7 revealed that such parameter has little effect in the

manganese adsorption for the range investigated as shown in Table IV.6. Therefore, it is

expected that manganese removal with bone char is not significantly affected by the

seasonal temperature changes in the region where AMD is generated.

Figure 4.7: Effect of temperature on the isotherm of manganese removal with bone char

from (a) laboratory solution (C0 = 100 mg L-1

), and (b) AMD effluent (150 min-1

, < 53

µm, 1/400 g mL-1

, 48 hours; continuous lines are Langmuir model).

0

5

10

15

20

25

0 10 20 30 40 50

q (

mg g

-1)

Ceq (mg L-1)

10°C

25°C

40°C

0

4

8

12

16

20

0 10 20 30 40 50

q (

mg g

-1)

Ceq (mg L-1)

10°C

25°C

40°C

(b)

(a)

47

Table IV.6: Parameters of Langmuir equation for the effect of temperature (150 min-1

,

pHi = 5.5-5.7, < 53 µm, 1/400 g mL-1

, 48 hours).

Temperature

(°C)

qm

(mg g-1

)

b x 10

(L mol-1

) R

2

Laboratory

solution

10 19 12.0 0.99

25 19 6.8 0.99

40 20 14.0 0.99

AMD

effluent

10 20 0.59 0.99

25 22 0.47 0.99

40 19 0.96 0.99

Gonçalves (2006) reported the use of bone char in alkaline medium for manganese

removal. 22% of manganese present in AMD effluent was removed using 0.2 g of bone

char and 200 mL of effluent for an initial concentration of 173 mg L-1

. The maximum

loading was 38 mg g-1

and was bigger than the maximum loading shown in Table IV.6.

The liquor obtained from the effect of temperature study at 25°C was also used to

evaluate the adsorption of fluoride species onto bone char. The results shown that after

contact with bone char, in the conditions describe above, the fluoride concentration in

the AMD effluent are in agreement with the Brazilian legislation and it was smaller than

the detection limit of the technique used for this analyze (<0.6 mg L-1

).

4.3.4. Mechanism for the removal of manganese with bone char

Cations removal from aqueous solutions by hydroxyapatite may generally occur due to

different sorption processes (adsorption, ion-exchange, surface complexation, co-

precipitation, recrystalization) depending on the experimental conditions and on the

nature of both sorbing cations and hydroxyapatite itself. Smiciklas et al. (2000) reported

that acidic as well as basic solutions in pH range 4–10 can be buffered after reaction

with the reactive surface sites of hydroxyapatite to its pHpzc value in which the surface

charge is zero.

48

According to Pan et al. (2009), the surface charge of the hydroxyapatite is predominated

by positively charged ≡ Ca 2+ and neutral ≡ P

0 species in acidic solutions, so the

surface charge of hydroxyapatite in this pH conditions is positive. On the other hand, in

al aline solutions, neutral ≡ Ca 0 and negatively charged ≡ P

− species may prevail

on the hydroxyapatite surface, so it becomes negatively charged:

≡ Ca 2+ ≡ Ca

0 + H

+ (4.3)

≡ P - + H

+ POH

0 (4.4)

The electrical characteristics of bone char was evaluated from measurements of zeta

potential by Rocha et al. (2011) and the results pointed out that the surface of bone char

is predominated by negatively charged in pH values above 5. This fact can increase the

electrostatic attraction forces between the bone char surface and the cations from the

solution, thus favoring the sorption of cations from aqueous solutions with pH > 5.

For investigate the release of calcium in solution, the bone char was washed as

described by Ribeiro (2011) in order to dissolve the calcite present in the material. The

calcium in solution (Figure 4.8) showed that ion exchange was involved in the removal

of manganese from AMD for this initial manganese concentration range. According to

’Connor and Weatherley (199 ), the hydroxyapatite component of the bone char plays

an important role in the adsorption of manganese, with adsorption occurring by ion

exchange. There was good linear relationship (R2 = 0.95) between the release of

calcium into solution and the manganese adsorbed to bone char.

49

Figure 4.8: Relationship between release of calcium into solution and manganese

adsorbed to bone char (25°C, 150 min-1

, 417–833 µm, 0.25/100 g mL-1

, pHi = 5.76, C0 =

100 mg L-1

, 24 hours).

The effect of diffusion as the rate controlling mechanism in the manganese removal by

bone char was evaluated according to the intraparticle diffusion model. It is known that

intraparticle diffusion plays a significant role in many adsorption processes and

normally the most significant contribution for the kinetic-mass transfer process is due to

the diffusion of adsorbate within the pore structure of the adsorbent. The intraparticle

diffusion model for adsorption is given by the following equation (Weber and Morris,

1963):

Ctkqpt

(4.5)

in which kp is the intraparticle diffusion constant which is evaluated as the slope of the

curve qt versus t1/2 and C is the constant related to the boundary layer thickness. If the

regression of qt vs. t1/2 is linear and passes through the origin, then intraparticle diffusion

is the sole rate-limiting step.

y = 1,292x - 0,0003 R² = 0,95

0,0000

0,0005

0,0010

0,0015

0,0020

0,0025

0,0030

0,0000 0,0004 0,0008 0,0012 0,0016 0,0020

Ca r

ele

ased (

mm

ol)

Mn adsorbed (mmol)

50

In order to evaluate the effect of intraparticle diffusion in the manganese adsorption

process, the qt was plotted as a function of square root of time, t1/2 at different

granulometry for laboratory solution and AMD effluent (see Figure 4.9). The plot for

intraparticle diffusion revealed a linear behavior passing near to the origin at the

beginning of manganese adsorption process (until 180 min for the particle size range of

<53 and 104-147 µm, and until 600 min for the particle size range of 417-833 µm). It

indicates that intraparticle diffusion is the main rate-limiting step for manganese

sorption systems but not the sole rate-limiting step since the plot for intraparticle

diffusion pass near to the origin. The kp values shown in Table IV.7 reveal that

intraparticle diffusion becomes slower when the particle size increases. This may occur

due to the fact that the particle size range of 417-833 µm has a major contribution of

smaller pores (Table IV.1) than the others particle size ranges studied in this work. It

may difficult the intraparticle diffusion of the metal, so manganese removal becomes

slower. The C constant values indicate that, the larger the intercept the greater the

boundary layer effect (Annadurai et al., 2008). Based on the results shown in Table

IV.7, the effect of external resistance increased when smaller particle size were used.

Table IV.7: Parameters of intraparticle diffusion model for manganese adsorption on

bone char (25°C, 150 min-1

, 2/400 g mL-1

, pHi = 5.5-5.7, C0 = 100 mg L-1

in the

laboratory solution).

Particle

size (µm)

kp

(mg g-1

min-0.5

)

C

(mg g-1

) R

2

Laboratory

solution

417-833 0.53 1.41 0.97

104-147 1.00 3.10 0.93

< 53 1.06 2.02 0.99

AMD

effluent

417-833 0.29 0.05 0.98

104-147 0.44 2.21 0.98

< 53 0.66 1.95 0.95

51

Figure 4.9: Intraparticle mass transfer for manganese removal with bone char from (a)

laboratory solution (C0 = 100 mg L-1

), and (b) AMD effluent (25°C, 150 min-1

, 2/400 g

mL-1

).

0

4

8

12

16

20

0 10 20 30 40 50 60

qt (m

g g

-1)

Time (min0.5)

417-833 µm

104-147 µm

<53 µm

0

3

6

9

12

15

0 10 20 30 40 50 60

qt (m

g.g

-1)

Time (min0.5)

417-833 µm

104-147 µm

<53 µm

(a)

(b)

52

4.4. Conclusions

In this study, batch equilibrium and kinetics tests were carried out to evaluate the

feasibility of using bone char for the removal of manganese from AMD effluents.

Experiments using a laboratory solution containing manganese at typical concentrations

were also done for comparative analysis. The following main conclusions can be drawn

based on the results obtained:

Equilibrium data was adequately described by the Langmuir model and the

maximum value of qm for manganese adsorption was 22 mg g-1

for AMD

effluent and 20 mg g-1

for laboratory solution;

Kinetic tests revealed that manganese removal by bone char is relatively slow

(equilibrium is reached after at least 600 minutes). It was verified also that

manganese loading follows a pseudo-second order model, thus indicating the

existence of a chemisorption mechanism probably involving an ion exchange

process between manganese and calcium ions;

The effect of competing species present in the AMD effluent for the bone char is

significant, so the pH of effluent must be previously raised to near neutral

condition to precipitate the other metals before contact with bone char. For

instance, at initial pH = 6.5, manganese removal of up to 70% was achieved,

while at initial pH = 5.0 the removal was lower than 50%.

The final concentration of fluoride and others metals present in the AMD

effluent was in agreement with the limit concentration established by Brazilian

legislation (CONAMA, 2005);

A relatively small effect of operating variables temperature and particle size was

verified for the investigated operating conditions, and;

The diffusion inside particles was identified as the main rate-limiting step for

manganese removal. However, when particles of smaller sizes are used, external

mass transfer and intraparticle diffusion contributions may affect manganese

removal.

Comparing with the current treatment of AMD effluents by precipitation with lime, a

smaller volume of sludge is expected. According to Carvalho et al. (2009), the

removal of manganese at pH close to the neutrality reduces by 50% the volume of

53

precipitate generated in the treatment of the effluent. In addition, the buffer effect of the

bone char may avoid the need for pH correction of effluent after treatment. Therefore,

the removal of manganese from AMD effluents using bone char as adsorbent was found

to be technically feasible and more efficient removal rates of manganese are expected

with the use of bone char in continuous column operations.

4.5. Acknowledgements

The authors acknowledge the financial support from FAPEMIG, CNPq, CAPES and

INCT-Acqua: National Institute of Science and Technology of Mineral Resources,

Water and Biodiversity. The authors also acknowledge the contribution from Bone Char

do Brasil Ltda as well as Prof. Sônia Denise Ferreira Rocha (DEMIN/UFMG) and Prof.

Ana Cláudia Ladeira (CDTN) for the fruitful discussion and contribution.

4.6. References

ANNADURAI, G.; LING, L.Y.; LEE, J.F. Adsorption of reactive dye from solution by

chitosan: isotherm, kinetic and thermodynamic analysis. Journal of Hazardous

Materials, v. 152, p. 337-346, 2008.

BAMFORTH, S.M.; MANNING, D.A.C.; SINGLETON, I.; YOUNGER, P.L.;

JOHNSON, K.L. Manganese removal from mine waters – investigating the occurrence

and importance of manganese carbonates. Applied Geochemistry, v. 21, p. 1274-1287,

2006.

BOSCO, S.M.D.; JIMENEZ, R.S.; CARVALHO, W.A. Removal of toxic metals from

wastewater by Brazilian natural scolecite. Journal of Colloid and Interface Science, v.

281, p. 424-431, 2005.

CARVALHO, G.X.; AGUIAR, A.O.; LADEIRA, A.C.Q. Estudo da remoção de

manganês de efluentes líquidos por precipitação. In: Proc. of the XXIII Encontro

Nacional de Tratamento de Minérios e Metalurgia Extrativa, Gramado, Brazil (in

Portuguese), 2009.

54

CHEN, Y.-N.; CHAI, L.-Y.; SHU, Y.-D. Study of arsenic(V) adsorption on bone char

from aqueous solution. Journal of Hazardous Materials, v. 160, p. 168-172, 2008.

CHEUNG, C.W.; PORTER, J.F.; McKAY, G. Sorption kinetics for the removal of

copper and zinc from effluents using bone char. Separation and Purification

Technology, v. 19, p. 55-64, 2000.

CHOY, K.K.H.; KO, D.C.K.; CHEUNG, C.W.; PORTER, J.F.; MCKAY, G. Film and

intraparticle mass transfer during the adsorption of metal ions onto bone char. Journal of

Colloid and Interface Science, v. 271, p. 284-295, 2004.

CONSELHO NACIONAL DO MEIO AMBIENTE. Resolução CONAMA n.º

357/2005: Classificação dos corpos de água e diretrizes ambientais para o seu

enquadramento, bem como estabelece as condições e padrões de lançamento de

efluentes. Brasília, 2005. Disponível na Web em:

http://www.mma.gov.br/port/conama/index.cfm.

GONÇALVES, C.R. Remoção de manganês e recuperação de urânio presentes em

águas ácidas de mina. Belo Horizonte, 2006. Dissertação (Mestrado) - Centro de

Desenvolvimento da Tecnologia Nuclear.

HASSAN, S.S.M; AWWAD, N.S.; ABOTERIKA, A.H.A. Removal of mercury(II)

from wastewater using camel bone charcoal. Journal of Hazardous Materials, v. 154, p.

992-997, 2008.

HO, Y.S. Review of second-order models for adsorption systems. Journal of Hazardous

Materials, B136, p. 681-689, 2006.

KORN, M.G.A., SANTOS JR., A.F., JAEGER, H.V., SILVA, N.M.S., COSTA, A.C.S.,

2004. Cooper, zinc and manganese determination in saline samples employing FAAS

after separation and preconcentration on Amberlite XAD-7 and Dowex 1X-8 loaded

with alizarin Red. Journal of Brazilian Chemical Society, v. 15(2), p. 212-218.

55

NASCIMENTO, M.R.L. Remoção e recuperaçâo de urânio de águas ácidas de mina por

resina de troca iônica. São Carlos, 1998. Dissertação (Mestrado) - Universidade Federal

de São Carlos.

PAN, X.; WANG, J.; ZHANG, D. Sorption of cobalt to bone char: kinetics, competitive

sorption and mechanism. Desalination, v. 249, p. 609-614, 2009.

ROBINSON-LORA, M.A.; BRENNAN, R.A. Biosorption of manganese onto chitin

and associated proteins during the treatment of mine impacted water. Chemical

Engineering Journal, v. 162, p. 565-572, 2010.

ROCHA, S.D.F.; RIBEIRO, M.V.; VIANA, P.R.M.; MANSUR, M.B. Bone char: an

alternative for removal of diverse organic and inorganic compounds from industrial

wastewater. In: Bhatnagar, A. (Org.), Application of adsorbents for water pollution,

Bentham Science Publishers Ltd., Ch. 14 (in press), 2011.

SMICIKLAS, I.D.; MILONJIC, S.K.; PFENDT, P.; RAICEVIC, S. The point of zero

charge and sorption of cadmium (II) and strontium (II) ions on synthetic hydroxyapatite.

Separation and Purification Technology, v. 18, p. 185-194, 2000.

SOTO, O.A.J.; TOREM, M.L.; TRINDADE, R.B.E. Palygorskite as a sorbent in the

removal of manganese(II) from water mine effluents. In: XIII INTERNATIONAL

CONFERENCE ON HEAVY METALS IN THE ENVIRONMENT, Rio de Janeiro -

RJ, 2005.

WEBER, W.J.; MORRIS, J.C. Kinetic of adsorption carbon solution, Journal Sanitary

Engineering Divisor, American Society Civil Engineers v. 89, p. 31-59, 1963 apud

ANNADURAI, G.; LING, L.Y.; LEE, J.F. Adsorption of reactive dye from solution by

chitosan: isotherm, kinetic and thermodynamic analysis. Journal of Hazardous

Materials, v. 152, p. 337-346, 2008.

56

5. ADSORPTION OF MANGANESE FROM ACID MINE DRAINAGE

EFFLUENTS USING BONE CHAR: CONTINUOUS FIXED BED COLUMN

AND BATCH DESORPTION STUDIES.

Dalila Chaves Sicupira a, Thiago Tolentino Silva

a, Ana Cláudia Queiroz Ladeira

b and

Marcelo Borges Mansur a,*

.

a Departamento de Engenharia Metalúrgica e de Materiais, Universidade Federal de

Minas Gerais

Av. Antônio Carlos, 6627, Campus Pampulha, 31270-901, Belo Horizonte, MG, Brazil

Tel.: +55 (31) 3409-1811; Fax: +55 (31) 3409-1716

E-mail: marcelo.mansur@demet.ufmg.br

* Corresponding author.

b Centro de Desenvolvimento da Tecnologia Nuclear, CNEN.

Av. Antônio Carlos, 6627, Campus Pampulha, 31270-901, Belo Horizonte, MG, Brazil

Abstract

In this study, continuous fixed bed column runs were done aiming to evaluate the

feasibility of using bone char for the removal of manganese from acid mine drainage

(AMD). Tests using laboratory solution containing solely manganese at typical

concentration level were also carried out for comparison purposes. The following

operating variables were evaluated: column height, flow rate and initial pH. Significant

variations in resistance to the mass transfer of manganese into the bone char were

identified using the Thomas model. A significant effect of bed height was observed only

in tests with laboratory solution. No significant change on the breakthrough volume was

observed with different flow rate. Increasing the initial pH from 2.96 to 5.50, the

breakthrough volume was increased. The maximum manganese loading capacity

in continuous tests using bone char for AMD effluent was 6.03 mg g-1

and for

laboratory solution was 26.74 mg g-1

. The study included also desorption tests using

solutions of HCl, H2SO4 and water aiming to reuse the adsorbent but no promising

results were obtained due to low desorption levels associated with relatively high mass

loss. Despite desorption results, the removal of manganese from AMD effluents using

(b)

57

bone char as adsorbent is technically feasible thus attending environmental legislation.

It is interesting to note that the use of bone char for manganese removal may avoid the

need for pH correction of effluent after treatment. Also, it can remove fluoride ions and

other metals, so bone char is a quite interesting alternative material for the treatment of

AMD effluents.

Key-words: manganese; bone char; acid mine drainage; continuous fixed bed column;

adsorption/desorption.

5.1. Introduction

Among the main environmental aspects and impacts of mining activities, the one

associated with the contamination of surface and groundwaters by acid mine drainage

(AMD) is probably the most significant one (Nascimento, 1998). AMD generation is

normally associated with the presence of sulfides like pyrite (FeS2), for instance, as

verified in the extraction of gold, coal, copper, zinc and uranium. Mining wastes when

exposed to oxidizing conditions in the presence of water may generate AMD, therefore

the proper disposal of mining wastes from such operations is crucial to prevent and/or

minimize its generation (Ladeira and Gonçalves, 2007). Besides waste rock piles, AMD

may also occur in open or underground pit mine, tailings storage and ore stockpiles. It

often contains high concentrations of SO42-

ions and the high acidity of such solutions

may promote the dissolution of metals like Zn, Cu, Cd, As, Fe, U, Al and Mn (Bamforth

et al., 2006; Robinson-Lora and Brennan, 2009). Such metal dissolution is probably the

most deleterious effect of AMD contamination because the resulting acidic and metal

contaminated streams can adversely impact humans and wildlife.

In Brazil, AMD has been highlighted in the mining region of Poços de Caldas, in the

state of Minas Gerais. The drainage generated in this region contains radionuclides (U,

Th and others) and species like Mn, Zn, Fe and F- ions at concentration levels above

those allowed by Brazilian legislation for direct discharge into the environment (Ladeira

and Gonçalves, 2007). The current treatment of such acid waters consists of metals

precipitation with lime followed by pH correction. Most of metal species are removed

but manganese ions removal from AMD is notoriously difficult to occur due to their

58

complex chemistry (Bamforth et al., 2006; Robinson-Lora and Brennan, 2010). For

complete precipitation of manganese, the pH must be raised to around 11, and such

operation involves a significant consumption of lime (Gonçalves, 2006). In addition,

after manganese removal, the pH must be neutralized for discharge so such treatment is

costly, generates large volumes of sludges and requires the consumption of high

quantities of reagents to be effective. New technologies to treat such effluents

containing manganese are then urged.

In this context, a promising method based on the removal of manganese using bone char

as adsorbent has been recently proposed (Sicupira et al., 2012). According to this

method, manganese was quantitatively removed from AMD at pH values near 6-7. As

an advantage, no pH correction of the treated effluent is necessary due to the buffer

effect of the bone char. Equilibrium and kinetics batch tests revealed that adsorption of

manganese with bone char was also influenced by the solid/liquid ratio. The particle

size and temperature studied had quite little effect on the manganese adsorption for the

operating range evaluated. The maximum value of qm found for manganese

adsorption based on Langmuir model was 22 mg g-1

.

In the view of the aforementioned, the aim of this work is to evaluate the feasibility of

treating AMD solutions containing manganese with bone char in continuous fixed bed

systems, in order to reach the levels required by the Brazilian legislation for direct

discharge of effluents. According to CONAMA (2005), the limit of total dissolved

manganese concentration in effluents is 1.0 mg L-1

and pH around 6-9. Fixed bed tests

were carried out using AMD effluents and laboratory solutions containing manganese at

typical concentrations for comparative analysis. Desorption studies were carried out

also with bone char previously loaded with manganese from laboratory solution in order

to evaluate the possibility to reuse it.

5.2. Experimental

59

5.2.1. Reagents

Bone char supplied by Bone Char do Brasil Ltda was used in this study. It contains

basically hydroxyapatite Ca10(PO4)6(OH)2 and calcite CaCO3 (Sicupira et al., 2012).

SEM-EDS analysis has shown a significant presence of calcium and phosphorus, as

expected; manganese and others metals were identified on its surface after contact with

AMD, thus suggesting that such metals were removed from the effluent possibly due to

some adsorptive and/or precipitating process. The main characteristics of the bone char

used in this study are as follows: real density = 2.9 g cm-3

, total pore volume = 0.275

cm3

g-1

, surface area = 93 m2

g-1

and particle size = 417-833 µm.

The AMD effluent used in this study was collected nearby one closed uranium mine in

Poços de Caldas region. Its metal composition is shown in Table V.1, which includes

the limit concentration for discharge in Brazil according to CONAMA (2005). It can be

seen that manganese concentration in the AMD exceeds the maximum value of

Brazilian legislation; the same occurs with zinc and acidity, but not to iron. After

precipitation with Ca(OH)2 30%, manganese concentration was still not in agreement

with legislation (see also in Table V.1), so additional treatment is required. Continuous

fixed bed column runs using a laboratory solution containing 100 mg L-1

of manganese

were also carried out for comparative analysis. Such solution was prepared using

chemicals of analytical reagent grade dissolved in distilled water.

60

Table V.1: Chemical characterization of the AMD effluent (Sicupira et al., 2012).

Parameters Concentration

(mg L-1

)

Concentration after

precipitation

(mg L-1

)

CONAMA 357

(mg L-1

)

U 6.8 <0.3 -

Mn 107.5 89.5 1.0

Ca 104.9 428 -

Mg 7.6 7.4 -

Al 164.2 28.5 -

Zn 17.7 4.0 5.0

Fe <0.01 <0.01 15.0

F- 99.0 38.0 10.0

SO42-

1349 1335 -

pH* 2.97 5.64 6 to 9

* All parameters are expressed in mg L-1

, except pH.

- Permissible level not defined by the Brazilian legislation.

5.2.2. Instrumentation

Atomic emission spectrometer with inductively coupled plasma (ICP-AES, Perkin

Elmer OPTIMA 7300DV) was used for assessing the concentration of manganese in the

aqueous solutions. Sulfate, phosphate and fluoride analysis were performed using a

DX500 Dionex ion chromatograph. The pH measurements were carried out using a pH

meter (Quimis). A scanning electron microscopy (SEM, SEI INSPECT S50) was

employed to analyze the surface morphology of the bone char.

5.2.3. Continuous fixed bed column runs

Continuous fixed bed column runs were conducted to determine the adsorption capacity

of bone char under continuous flow conditions. All column runs were performed in

down flow mode. The column was filled with 40 g of bone char with particle size of

417-833 µm. The effluent (pHi = 2.96 or 5.50) was continuously fed into the column

using a peristaltic pump (Cole-Parmer, Masterflex); in the case of laboratory solution

61

the pHi = 5.76 was used. The range of initial pH was chosen based on previous batch

equilibrium and kinetics studies (Sicupira et al., 2012); the AMD effluent sample has a

typical pH value of approximately 3. The diameter of the adsorptive bed of the column

was 2.2 cm and a hydraulic flow rate of 7.5 mL min-1

was maintained constant. Effluent

samples were collected at determined contact times from the column, filtered and

acidified using HNO3 before being analyzed by ICP-AES. The fixed bed studies were

also carried out with flow rate of 3.0 mL min-1

to evaluate the effect of the flow rate in

the fixed bed column and the effect of the mass of bone char was also evaluated using

20 g of bone char. As the temperature did not show significant effect on the adsorption

of manganese onto bone char in batch studies (Sicupira et al., 2012), all continuous

runs were carried out at room temperature. Breakthrough curves were plotted showing

the concentration ratio Ct/C0 (manganese concentration at time t divided by the initial

concentration of manganese) versus the number of bed volumes (BV), which is defined

as the eluted volume divided by the volume of bed’s voids (Cussler, 199 ).

5.2.4. Modeling

The Thomas model was chosen to evaluate the adsorption characteristics of manganese

from AMD effluents in continuous column operation due to its simplicity. It is widely

used in the modeling of fixed bed breakthrough curves (Unuabonah et al., 2010). The

model assumes a Langmuir (favorable) isotherm in accordance to previous investigation

on the adsorption of manganese onto bone char (Sicupira et al., 2012) and it is given by

the following equation (Chu, 2010):

Ct

C0

1

1 exp q0

C0t

(5.1)

The linearized form of Thomas model can be expressed as follows:

n C0

Ct

1 q0

C0 (5.2)

in which kT (L mg-1

min-1

) is the Thomas rate constant, q0 is the sorption capacity of the

adsorbent per unit mass (mg g-1

), M is the mass of adsorbent (g) and Q is the flow rate

62

(L min-1

). The kinetic coefficient kT and the adsorption capacity q0 can be obtained

from the plot of ln C0

Ct

1 against time (t) at a given flow rate using linear regression

(Cavas et al., 2011).

5.2.5. Desorption studies

Metal desorption was evaluated using selected eluents: HCl (0.01M, 0.001M,

0.0001M), H2SO4 (0.01M, 0.001M, 0.0001M) and H2O. Desorption was carried out by

contacting a loaded sorbent (500 mg, dry weight, previously saturated with metals,

obtained from batch adsorption experiments) with a given volume of eluent (100 mL)

under agitation (150 rpm) at 25ºC using a mechanical shaker (INNOVA 44). Samples (1

mL) were collected at fixed contact times, diluted and filtered before being analyzed by

ICP-AES.

5.3. Results and Discussion

5.3.1 Fixed bed sorption runs

The effect of bed height on the sorption of manganese in the continuous fixed bed

column is shown in Figure 5.1. Analyzing the curves obtained, the breakthrough curves

for the laboratory solution did not show the typical “ ” shape (see Figure 5.1a) as

verified in runs with AMD effluent (see Figure 5.1b). In fact, the curves obtained in the

runs with laboratory solutions are more inclined to the right. Such behavior is an

indicative that the mass-transfer zone is almost as long as the bed and that significant

resistance are involved, so a longer period of time is necessary to reach the

saturation point (Quek and Al-Duri, 2007) and less than one-half of the bed capacity is

utilized (McCabe et al., 2005). The difference in the shape of the curves is due to the

distinct flow rates used in the runs. According to Gonçalves (2006), the breakthrough

curve is sharper with the increasing of the adsorption rate, i.e., at lower flow rates and

lower initial concentration of feed solution.

63

Figure 5.1: Effect of bed height on the breakthrough of manganese removal with bone

char in a continuous fixed bed column (417-833 µm): (a) Laboratory solution (flow rate

= 7.5 mL min-1

, C0 = 100 mg L-1

, pHi = 5.76), (b) AMD effluent (flow rate = 3.0 mL

min-1

, pHi = 2.96, continuous curves are Thomas model).

0,0

0,2

0,4

0,6

0,8

1,0

1,2

0 100 200 300 400 500 600

C/C

0

Bed Volume (BV)

20 g

40 g

0,0

0,2

0,4

0,6

0,8

1,0

1,2

0 10 20 30 40 50 60 70 80

C/C

0

Bed volume (BV)

20 g

40 g

(a)

(b)

64

The constants of Thomas model (kT and q0) and operating parameters BVb

(breakthrough point that refers to the maximum manganese concentration in effluents

limited by CONAMA at 1.0 mg L-1

) and BVs (saturation point that refers to final

manganese concentration is equal to the initial manganese concentration) for laboratory

solution and AMD effluent are shown in Table V.2. Results for laboratory solution

reveal that an increase in the mass of bone char increases the breakthrough point.

Increasing the mass of bone char from 20 g to 40 g, at a constant flow rate, the

breakthrough time was extended from 50 min to 280 min and the number BVb was

extended from 11.70 to 34.16 due to the increasing in the number of binding sites. In the

case of AMD effluent, increasing the mass of bone char from 20 g to 40 g, at a constant

flow rate, the breakthrough time was extended from 60 min to 120 min, but the number

BVb was unchanged; in fact, the same value of 5.85 for both mass of bone char was

obtained and it was smaller than the BVb values obtained for the laboratory solution.

This can be attributed to the fact that, in the case of AMD effluent (pHi = 2.96), the

initial pH is lower and that there are others species in solution that may limit the

removal of manganese. A precipitate formed on the surface of the loaded bone char can

be seen in Figure 5.2 that probably blocked the binding sites, preventing the progressing

of the removal process. In the case of AMD effluent, is also interesting to note that the

pH of the effluent that leave the column dropped quickly thus reducing manganese

adsorption.

Concerning the values of rate constant of Thomas model (kT) and the maximum loads

(q0) shown in Table V.2, quite different values were obtained for kT for both solutions,

which means there are significant variations in resistance to the mass transfer into the

bone char for both systems. The q0 values are bigger for the longer bed height in the

laboratory solution due to the increasing in the number of binding sites. For AMD

effluent, the q0 values are approximately the same, since the manganese removal was

limited by the precipitate formed on the bone char surface and the low pH value of the

effluent that leave the column and not by the binding sites. As can be seen by

comparing Figure 5.1 and Table V.2, the higher the value of the constant kT, the more

effective is the mass transfer process, thus resulting on sharper “ ” curves.

65

Table V.2: Parameters of Thomas model for the adsorption of manganese onto bone

char (417-833 µm).

Mass of

bone

char (g)

Flow rate

(mL min-

1)

initial

pH

final

pH

BVb BVs q0

(mg g-1

)

kT x10

4

(mg g-1

min-1

)

R2

Laboratory

solution

20

20

40

3.0

7.5

7.5

5.76

5.76

5.76

7.34

7.34

7.34

11.70

11.70

34.16

444.60

444.60

556.35

15.97

15.85

26.74

0.23

0.44

0.24

0.95

0.95

0.99

AMD

effluent

20

20

40

3.0

3.0

3.0

2.96

5.50

2.96

4.45

5.55

4.45

5.85

17.55

5.85

35.09

70.19

35.09

2.85

6.03

2.60

3.00

1.32

1.58

0.97

0.97

0.99

66

Figure 5.2: Morphology of bone char by SEM (scanning electron microscopy): (a) bone

char as received, and (b) bone char after contact with AMD effluent in the column.

(a)

(b)

67

The effect of flow rate on the manganese removal in the continuous fixed bed column is

shown in Figure 5.3. No significant change on the breakthrough point was verified with

the increase in the evaluated flow rate range. As expected, faster saturation of the bed

was obtained with the increase of the flow rate from 3.0 mL min-1

to 7.5 mL min-1

, at a

same bed height. The BVb was also the same value of 11.70 because it is limited by the

number of binding sites.

Figure 5.3: Effect of flow rate on the breakthrough of manganese removal with bone

char in a continuous fixed bed column using the laboratory solution (C0 = 100 mg L-1

,

pHi = 5.76, mass = 20 g, 417-833 µm).

The effect of the initial pH of AMD effluent was also evaluated and results are shown in

Figure 5.4. Increasing the initial pH from 2.96 to 5.50, at the same operating conditions,

is possible to observe that the saturation point is quickly reached for the pHi = 2.96, due

to the fact that the pH value of the effluent that leaves the column dropped from

approximately 8.0 to 2.96 in a few minutes. The saturation point occurred after 240 min

for pHi = 2.96 and after 660 min for pHi = 5.50. As observed by Sicupira et al. (2012),

manganese adsorption does not occur in low pH values. In fact, increasing the initial pH

from 2.96 to 5.50, at a constant flow rate, the breakthrough time was extended from 360

0,0

0,2

0,4

0,6

0,8

1,0

0 100 200 300 400 500

C/C

0

Bed Volume (BV)

3.0 mL/min

7.5 mL/min

68

min to 720 min, and the number of BVb was extended from 35.09 to 70.19. Regarding

the values of Thomas model shown in Table V.2, the value obtained to q0 is bigger for

pHi = 5.50 than for pHi = 2.96, since the manganese adsorption is favored by the

progressing of the pH value during the experiment at pHi = 5.50. Moreover, the AMD

effluent at pHi = 2.96, has other metals ions in solution that compete with manganese for

binding sites. This can occur or by the competition of the other metals ions with

manganese or by formation of precipitate of these metals ions on the bone char surface

that can block the binding sites. As no work was found in the literature using bone char

as adsorbent for the removal of manganese in fixed bed columns, no comparison of the

results obtained in this study was possible. The maximum manganese loading calculated

in continuous tests using bone char was 6.03 mg g-1

for AMD effluent and 26.74 mg g-1

for laboratory solution. The value for laboratory solution is larger than that obtained in

batch tests, 20 mg g-1

, as expected (Guimarães, 2010), but in the case of AMD effluent

it was not verified. It can be explain by the progressing of pH value along the column

when AMD effluent is fed. Every moment a new solution is fed into the column and the

calcite present in the bone char is rapidly consumed. This make the pH value of the

effluent that leaves the column dropped from approximately 8.0 to 2.96 in a few

minutes. Another fact that can also prevent the manganese removal in fixed bed is the

high concentration of calcium in the effluent resulted from the precipitation with

Ca(OH)2. According to Sicupira et al. (2012), the mechanism of manganese removal is

based on the ion exchange with the calcium present in the structure of hydroxyapatite,

so high concentrations of calcium in the effluent may affect negatively the ion exchange

of calcium with manganese.

69

Figure 5.4: Effect of initial pH on the breakthrough of manganese removal with bone

char in a continuous fixed bed column using the AMD effluent

(flow rate = 3.0 mL min-1

, mass = 20 g, 417-833 µm, continuous curves are Thomas

model).

5.3.2 Desorption studies

The design of adsorption processes requires desorption studies to be fully implemented.

When the adsorbent becomes exhausted, the recovery of the loaded adsorbent may be

advantageous from the economic point of view as much as it can be reused in closed

circuit. In fact, regeneration of spent adsorbent columns is quite an important process in

wastewater treatment (Mohan and Chander, 2006).

Previous study revealed that manganese adsorption onto bone char occurs

as chemisorption (Sicupira et al., 2012). As chemisorption usually results in more stable

adsorbed species, desorption is expected to be more difficult to occur. Based on

the results shown in Table V.3, the manganese desorption from bone char seems to be

not interesting for reusing the material. Using HCl as eluent, the highest desorption

obtained was 22.4% with mass loss of 38.8% during 15 minutes of agitation. Using the

same conditions but for 5 minutes of agitation, desorption was 21.8% with mass loss of

0,0

0,2

0,4

0,6

0,8

1,0

0 30 60 90 120 150 180

C/C

0

Bed volume (BV)

pHi=2.96

pHi=5.50

70

1.8%. Using H2SO4 as eluent, the better result was obtained for H2SO4 0.01M with

45.5% of desorption but 36.1% of mass loss after 15 minutes of agitation. Regarding

these results, it can be observed that manganese desorption from bone char is not viable

due to low desorption levels associated with relatively high mass losses which can even

modify the characteristics of the bone char.

Table V.3: Desorption of manganese from loaded bone char using various elution

solutions (500 mg, 100 mL, 25°C, 150 rpm, 417-833 µm).

Eluent Eluent Concentration

(mol L-1

)

Desorption of

manganese(%)

Mass loss

(%)

HCl

10-2

21.8* 1.8*

10-2

22.4 38.8

10-3

6.3 3.0

10-4

0.0 0.0

H2SO4

10-2

45.5 36.1

10-3

5.2 3.1

10-4

0.0 0.0

H2O - 0.0 0.0

* t = 5 min.

Desorption experiments with bone char loaded with zinc were carried out by Guedes et

al. (2007). It was observed that desorption with distilled water was not effective but in

the case of desorption carried out with the sulphuric acid solution (0.01 mol L-1

), the

levels of desorption was 40% which can be compared with the level of manganese

desorption shown in Table V.3 for the same eluent solution that was of 45%. The

authors cited did not report the adsorbent mass loss and desorption experiments with

bone char loaded with Mn were not reported.

5.4. Conclusions

In this study, fixed bed tests were carried out to evaluate the feasibility of using bone

char for the removal of manganese from AMD effluents in order to reach the maximum

71

concentration permissible by Brazilian legislation (CONAMA, 2005). Experiments

using a laboratory solution containing manganese at typical concentrations were also

done for comparative analysis. The following main conclusions can be drawn based on

the results obtained:

For laboratory solution an increase in the mass of bone char increased the

number of BVb which was extended from 11.70 to 34.16. In the case of AMD

effluent, increasing the mass of bone char, the BVb was unchanged due to the

other species in solution and/or the precipitate formed on the surface which may

limit the removal of manganese.

Concerning the values of rate constant of Thomas model (kT) it can be

concluded that there are significant variations in resistance to the mass transfer

into the bone char for laboratory solutions and AMD effluent. For laboratory

solution the mass-transfer zone was almost as long as the bed, so a longer period

of time is necessary to reach the saturation point.

Regarding the influence of flow rate, it was observed that there was no

significant change on the number of BVb. With the increase of the flow rate, the

bed was saturated faster. The BVb was also the same value of 11.70 because it is

limited by the number of binding sites.

Increasing the initial pH of AMD effluent from 2.96 to 5.50, the BVb was

extended from 35.09 to 70.19, as expected.

The maximum manganese loading calculated in continuous tests using bone

char for AMD effluent was 6.03 mg g-1

and for laboratory solution was 26.74

mg g-1

.

Manganese desorption tests indicated that reuse of bone char seems to be not

applicable. The highest desorption was obtained for H2SO4 0.01M with 45.5%

of desorption, but 36.1% of mass loss after 15 minutes of agitation. Manganese

desorption from bone char is not feasible due to low desorption levels associated

with relatively high mass loss which can even modify the characteristics of the

bone char.

The removal of manganese from AMD effluents using bone char as adsorbent is

technically feasible thus attending environmental legislation, despite desorption seems

72

not applicable for the eluents solutions investigated. It is interesting to note that bone

char may avoid the need for pH correction of effluent after treatment as currently done

in the treatment with lime. Also, it can remove fluoride ions and other metals present in

AMD effluent (Sicupira et al., 2012), so bone char is a quite interesting alternative

material for the treatment of AMD effluents.

5.5. Acknowledgments

The authors acknowledge the financial support from FAPEMIG, CNPq, CAPES and

INCT-Acqua: National Institute of Science and Technology of Mineral Resources,

Water and Biodiversity. The contributions from Bone Char do Brasil Ltda and INB

(Indústrias Nucleares do Brasil) as well as the fruitful discussions with Prof. Sônia

Denise Ferreira Rocha (DEMIN/UFMG) are also kindly acknowledged.

5.6. References

BAMFORTH, S.M.; MANNING, D.A.C.; SINGLETON, I.; YOUNGER, P.L.;

JOHNSON, K.L. Manganese removal from mine waters – investigating the occurrence

and importance of manganese carbonates. Applied Geochemistry, v. 21, p. 1274-1287,

2006.

CAVAS, L.; KARABAYA, Z.; ALYURUKA, H.; DOGAN, H.; DEMIR, G.K. Thomas

and artificial neural network models for the fixed-bed adsorption of methylene blue by a

beach waste Posidonia oceanica (L.) dead leaves. Chemical Engineering Journal, v.

171, p. 557-562, 2011.

CHU, K.H. Fixed bed sorption: Setting the record straight on the Bohart–Adams and

Thomas models. Journal of Hazardous Materials, v. 177, p. 1006-1012, 2010.

CONSELHO NACIONAL DO MEIO AMBIENTE. Resolução CONAMA n.º

357/2005: Classificação dos corpos de água e diretrizes ambientais para o seu

enquadramento, bem como estabelece as condições e padrões de lançamento de

efluentes. Brasília, 2005. Disponível na Web em:

http://www.mma.gov.br/port/conama/index.cfm.

73

CUSSLER, E.L. Diffusion: Mass Transfer in Fluid Systems. 2nd edition. Cambridge

University Press, p. 312-314, 1997.

GONÇALVES, C.R. Remoção de manganês e recuperação de urânio presentes em

águas ácidas de mina. Belo Horizonte, 2006. Dissertação (Mestrado) - Centro de

Desenvolvimento da Tecnologia Nuclear.

GUEDES, T.S.; MANSUR, M.B.; ROCHA, S.D.F. A perspective of bone char use in

the treatment of industrial liquid effluents containing heavy metals. In: XXI ENTMME,

Ouro Preto - MG, 2007.

GUIMARÃES, D. Tratamento de efluentes ricos em sulfato por adsorção em resinas de

troca iônica. Ouro Preto, 2010. Dissertação (Mestrado) - UFOP.

LADEIRA, A.C.Q.; GONÇALVES, C.R. Influence of anionic species on uranium

separation from acid mine water using strong base resins. Journal of Hazardous

Materials, v. 148, p. 499–504, 2007.

McCABE, W.L.; SMITH, J.C.; HARRIOTT, P. Unit operations of chemical

engineering: Chemical Engineering Series. 7th edition. McGraw-Hill´s Science, p. 836-

847, 2005.

MOHAN, D.; CHANDER, S. Removal and recovery of metal ions from acid mine

drainage using lignite - A low cost sorbent. Journal of Hazardous Materials, v. B137, p.

1545-1553, 2006.

NASCIMENTO, M.R.L. Remoção e recuperaçâo de urânio de águas ácidas de mina por

resina de troca iônica. São Carlos, 1998. Dissertação (Mestrado) - Universidade Federal

de São Carlos.

74

QUEK, S.Y.; AL-DURI, B. Application of film-pore diffusion model for the adsorption

of metal ions on coir in fixed bed column. Chemical Engineering and Processing, v. 46,

p. 477-485, 2007.

ROBINSON-LORA, M.A.; BRENNAN, R.A. Biosorption of manganese onto chitin

and associated proteins during the treatment of mine impacted water. Chemical

Engineering Journal. v. 162, p. 565-572, 2010.

SICUPIRA, D.C.; SILVA, T.T.; MANSUR, M.B. Batch removal of manganese from

acid mine drainage using bone char. Hydrometallurgy, submitted, 2012.

UNUABONAH, E.I.; OLU-OWOLABI, B.I.; FASUYI, E.I.; ADEBOWALE, K.O.

Modeling of fixed-bed column studies for the adsorption of cadmium onto novel

polymer–clay composite adsorbent. Journal of Hazardous Materials, v. 179, p. 415-

423, 2010.

75

6. FINAL CONSIDERATIONS AND SUGGESTIONS FOR FUTURE WORKS

In this study, batch and continuous tests were carried out to evaluate the feasibility of

using bone char as an alternative material for the removal of manganese from AMD

effluents. Experiments using a laboratory solution containing manganese at typical

concentrations were also done for comparative analysis. The maximum value of qm for

manganese adsorption based on Langmuir model was 22 mg g-1

for batch tests and 6.03

mg g-1

for continuous tests. Kinetic tests revealed that manganese removal by bone char

is relatively slow and has a relatively small effect of operating variables temperature

and particle size. The effect of competing species present in the AMD effluent for the

bone char was significant, so the pH of effluent must be raised to near neutral

conditions to precipitate the others metals before contact with bone char. Comparing

with the current treatment of AMD effluents by precipitation with lime, a smaller

volume of sludge is expected. According to Carvalho et al. (2009), the removal of

manganese at pH close to the neutrality reduces by 50% the volume of

precipitate generated in the treatment of the effluent. The removal of manganese from

AMD effluents using bone char as adsorbent was technically feasible, but the removal

rates of manganese need to be increasing for to be applied in the industry. However, it is

interesting to note that the bone char may avoid the need for pH correction of effluent

after treatment and can also remove the fluoride and the others metals present in AMD

effluent. Manganese desorption from bone char is not applicable due to low desorption

levels associated with relatively high mass loss which can even modify the

characteristics of the bone char.

On the regard to such final considerations and given the results obtained so far with this

work, the following aspects are suggested for future investigations:

1) Evaluate the use of another precipitating agent instead of Ca(OH)2 to increase the pH

of the effluent and do a comparative analysis.

2) Evaluate the performance of bone char at manganese adsorption in continuous tests

with the effluent pHi= 7.0.

76

3) Investigate the chemical characterization of bone char after adsorption in fixed-bed

systems in order to identify the compounds formed on the surface that blocked the

binding sites.

4) Monitor, during the column experiments, the sorption of the sulfate and fluoride

anions onto bone char.

5) Perform a comparative economic evaluation between current treatment of AMD with

lime and bone char in columns.

6) Quantify the contribution of the external mass transfer effect on the manganese

removal at changing operating conditions.